From owner-freebsd-hardware  Sun Feb 23 05:59:48 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id FAA10956
          for hardware-outgoing; Sun, 23 Feb 1997 05:59:48 -0800 (PST)
Received: from yoda.fdt.net (root@yoda.fdt.net [205.229.48.17])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA10943
          for <hardware@freebsd.org>; Sun, 23 Feb 1997 05:59:46 -0800 (PST)
Received: from Kryten.nina.org (port-25.ts1.gnv.fdt.net [205.229.51.25]) by yoda.fdt.net with SMTP id IAA24356; Sun, 23 Feb 1997 08:52:54 -0500
Date: Sun, 23 Feb 1997 08:52:42 -0500 (EST)
From: Frank Seltzer <frankd@yoda.fdt.net>
X-Sender: frankd@Kryten.nina.org
To: Bruce Evans <bde@zeta.org.au>
cc: bkogawa@primenet.com, dkelly@hiwaay.net, hardware@freebsd.org
Subject: Re: _big_ IDE disks?
In-Reply-To: <199702210824.TAA01446@godzilla.zeta.org.au>
Message-ID: <Pine.BSF.3.95.970223084958.9399C-100000@Kryten.nina.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Fri, 21 Feb 1997, Bruce Evans wrote:

> The fast-i586-copyout option is not enabled in 2.2-GAMMA.  I forget what
> it was in ALPHA.  It was either not enabled or buggy (it caused panics).

Is there a way to enable this in 2.2-GAMMA? This is what I am seeing here:

Kryten /usr/ports/distfiles % dd if=/dev/zero of=/dev/null bs=1m
count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)

Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?

Frank
--
Only in America can a homeless veteran sleep in a cardboard box while a 
draft dodger sleeps in the White House - anonymous


From owner-freebsd-hardware  Sun Feb 23 06:48:54 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id GAA13507
          for hardware-outgoing; Sun, 23 Feb 1997 06:48:54 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA13502
          for <hardware@FreeBSD.ORG>; Sun, 23 Feb 1997 06:48:50 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id BAA01194; Mon, 24 Feb 1997 01:41:42 +1100
Date: Mon, 24 Feb 1997 01:41:42 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702231441.BAA01194@godzilla.zeta.org.au>
To: bde@zeta.org.au, frankd@yoda.fdt.net
Subject: Re: _big_ IDE disks?
Cc: bkogawa@primenet.com, dkelly@hiwaay.net, hardware@FreeBSD.ORG
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>> The fast-i586-copyout option is not enabled in 2.2-GAMMA.  I forget what
>> it was in ALPHA.  It was either not enabled or buggy (it caused panics).
>
>Is there a way to enable this in 2.2-GAMMA? This is what I am seeing here:

Update to the 2.2 npx.c.

>Kryten /usr/ports/distfiles % dd if=/dev/zero of=/dev/null bs=1m
>count=1000
>1000+0 records in
>1000+0 records out
>1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)
>
>Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?

They seem a bit low.  I would hope for 80-90MB/sec.  The fast-i586-copyout
wouldn't affect this because it is an option for i586's.

Bruce

From owner-freebsd-hardware  Sun Feb 23 07:11:51 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id HAA14760
          for hardware-outgoing; Sun, 23 Feb 1997 07:11:51 -0800 (PST)
Received: from yoda.fdt.net (root@yoda.fdt.net [205.229.48.17])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA14755
          for <hardware@FreeBSD.ORG>; Sun, 23 Feb 1997 07:11:47 -0800 (PST)
Received: from Kryten.nina.org (port-25.ts1.gnv.fdt.net [205.229.51.25]) by yoda.fdt.net with SMTP id KAA27263; Sun, 23 Feb 1997 10:11:10 -0500
Date: Sun, 23 Feb 1997 10:11:40 -0500 (EST)
From: Frank Seltzer <frankd@yoda.fdt.net>
X-Sender: frankd@Kryten.nina.org
To: Bruce Evans <bde@zeta.org.au>
cc: bkogawa@primenet.com, dkelly@hiwaay.net, hardware@FreeBSD.ORG
Subject: Re: _big_ IDE disks?
In-Reply-To: <199702231441.BAA01194@godzilla.zeta.org.au>
Message-ID: <Pine.BSF.3.95.970223095547.9399D-100000@Kryten.nina.org>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Mon, 24 Feb 1997, Bruce Evans wrote:

> >> The fast-i586-copyout option is not enabled in 2.2-GAMMA.  I forget what
> >> it was in ALPHA.  It was either not enabled or buggy (it caused panics).
> >
> >Is there a way to enable this in 2.2-GAMMA? This is what I am seeing here:
> 
> Update to the 2.2 npx.c.

I am running 2.2-GAMMA from Feb 21. Does this version have this npx.c? If
not, where can I get it?

> >Kryten /usr/ports/distfiles % dd if=/dev/zero of=/dev/null bs=1m
> >count=1000
> >1000+0 records in
> >1000+0 records out
> >1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)
> >
> >Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?
> 
> They seem a bit low.  I would hope for 80-90MB/sec.  The fast-i586-copyout
> wouldn't affect this because it is an option for i586's.

Would any BIOS settings have an effect on this? For example, I have video
BIOS shadow enabled but not the individual memory address settings. This
is on an Asus P6NP5 motherboard.

> Bruce
> 

Frank
--
Only in America can a homeless veteran sleep in a cardboard box while a 
draft dodger sleeps in the White House - anonymous


From owner-freebsd-hardware  Sun Feb 23 07:21:21 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id HAA15187
          for hardware-outgoing; Sun, 23 Feb 1997 07:21:21 -0800 (PST)
Received: from mexico.brainstorm.eu.org (root@mexico.brainstorm.fr [193.56.58.253])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA15181
          for <hardware@freebsd.org>; Sun, 23 Feb 1997 07:21:16 -0800 (PST)
Received: from brasil.brainstorm.eu.org (brasil.brainstorm.fr [193.56.58.33])
          by mexico.brainstorm.eu.org (8.8.4/8.8.4) with ESMTP
	  id QAA00842 for <hardware@freebsd.org>; Sun, 23 Feb 1997 16:21:07 +0100
Received: (from uucp@localhost) by brasil.brainstorm.eu.org (8.8.4/8.6.12) with UUCP id QAA13522 for hardware@freebsd.org; Sun, 23 Feb 1997 16:21:03 +0100
Received: (from roberto@localhost) by keltia.freenix.fr (8.8.5/keltia-uucp-2.9) id QAA28449;
          Sun, 23 Feb 1997 16:20:45 +0100 (CET)
Message-ID: <19970223162044.MY34128@keltia.freenix.fr>
Date: Sun, 23 Feb 1997 16:20:44 +0100
From: roberto@keltia.freenix.fr (Ollivier Robert)
To: hardware@freebsd.org
Subject: Re: _big_ IDE disks?
References: <199702210824.TAA01446@godzilla.zeta.org.au> <Pine.BSF.3.95.970223084958.9399C-100000@Kryten.nina.org>
X-Mailer: Mutt 0.60,1-3,9
Mime-Version: 1.0
X-Operating-System: FreeBSD 3.0-CURRENT ctm#2999
In-Reply-To: <Pine.BSF.3.95.970223084958.9399C-100000@Kryten.nina.org>; from Frank Seltzer on Feb 23, 1997 08:52:42 -0500
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

According to Frank Seltzer:
> 1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)
> 
> Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?

Not really.

202 [16:19] roberto@caerdonn:~> dd if=/dev/zero of=/dev/null bs=1m
count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 11.455249 secs (91536726 bytes/sec)

HP XU with P6/180, 64 MB EDO.
-- 
Ollivier ROBERT    -=- The daemon is FREE! -=-    roberto@keltia.freenix.fr
  FreeBSD keltia.freenix.fr 3.0-CURRENT #39: Sun Feb  2 22:12:44 CET 1997

From owner-freebsd-hardware  Sun Feb 23 08:29:46 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA17325
          for hardware-outgoing; Sun, 23 Feb 1997 08:29:46 -0800 (PST)
Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA17320
          for <hardware@freebsd.org>; Sun, 23 Feb 1997 08:29:43 -0800 (PST)
Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA11518; Sun, 23 Feb 1997 09:29:36 -0700 (MST)
Date: Sun, 23 Feb 1997 09:29:36 -0700 (MST)
Message-Id: <199702231629.JAA11518@rocky.mt.sri.com>
From: Nate Williams <nate@mt.sri.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Bruce Evans <bde@zeta.org.au>
Cc: hardware@freebsd.org
Subject: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <199702231441.BAA01194@godzilla.zeta.org.au>
References: <199702231441.BAA01194@godzilla.zeta.org.au>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> >> The fast-i586-copyout option is not enabled in 2.2-GAMMA.  I forget what
> >> it was in ALPHA.  It was either not enabled or buggy (it caused panics).
> >
> >1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)
> >
> >Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?
> 
> They seem a bit low.  I would hope for 80-90MB/sec.  The fast-i586-copyout
> wouldn't affect this because it is an option for i586's.

Assuming that I'm running the correct code:

FreeBSD 2.2-GAMMA #10: Fri Feb 21 18:07:37 MST 1997
CPU: Pentium (99.47-MHz 586-class CPU)
  Origin = "GenuineIntel"  Id = 0x525  Stepping=5
  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 67108864 (65536K bytes)
avail memory = 63631360 (62140K bytes)

 *      $Id: npx.c,v 1.31.2.6 1997/02/13 06:59:51 bde Exp $

moth:/sys/i386/isa # dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 14.549281 secs (72070640 bytes/sec)

Pretty low.  This is an ASUS/Triton-I board if it makes any difference.


Nate

From owner-freebsd-hardware  Sun Feb 23 17:13:57 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id RAA15122
          for hardware-outgoing; Sun, 23 Feb 1997 17:13:57 -0800 (PST)
Received: from gargoyle.bazzle.com (gargoyle.bazzle.com [206.103.246.190])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA15117
          for <hardware@freebsd.org>; Sun, 23 Feb 1997 17:13:51 -0800 (PST)
Received: from gargoyle.bazzle.com (net2.bazzle.com [206.103.246.189]) by gargoyle.bazzle.com (8.8.5/8.6.12) with SMTP id UAA09362; Sun, 23 Feb 1997 20:13:43 -0500 (EST)
Date: Sun, 23 Feb 1997 20:13:43 -0500 (EST)
From: "Eric J. Chet" <ejc@gargoyle.bazzle.com>
To: Ollivier Robert <roberto@keltia.freenix.fr>
cc: hardware@freebsd.org
Subject: Re: _big_ IDE disks?
In-Reply-To: <19970223162044.MY34128@keltia.freenix.fr>
Message-ID: <Pine.BSF.3.95.970223200907.9235B-100000@gargoyle.bazzle.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Sun, 23 Feb 1997, Ollivier Robert wrote:

> According to Frank Seltzer:
> > 1048576000 bytes transferred in 16.608622 secs (63134437 bytes/sec)
> > 
> > Do these numbers seem reasonable for a P6 200 w/64 megs EDO RAM?
> 
> Not really.
> 
> 202 [16:19] roberto@caerdonn:~> dd if=/dev/zero of=/dev/null bs=1m
> count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 11.455249 secs (91536726 bytes/sec)
> 
> HP XU with P6/180, 64 MB EDO.
> -- 
> Ollivier ROBERT    -=- The daemon is FREE! -=-    roberto@keltia.freenix.fr
>   FreeBSD keltia.freenix.fr 3.0-CURRENT #39: Sun Feb  2 22:12:44 CET 1997
> 
Hello

FreeBSD 3.0-CURRENT #0: Fri Feb 21 22:59:34 EST 1997
    ejc@gargoyle.bazzle.com:/var/usr/src/sys/compile/gargoyle
Calibrating clock(s) relative to mc146818A clock ... i586 clock: 133673485
Hz, i
8254 clock: 1193515 Hz
CPU: Pentium (133.64-MHz 586-class CPU)
  Origin = "GenuineIntel"  Id = 0x526  Stepping=6
  Features=0x3bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8,APIC>

This is a GigaByte 586DX-dual pentium board.
# dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 9.184039 secs (114173731 bytes/sec)

Eric


From owner-freebsd-hardware  Sun Feb 23 17:17:13 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id RAA15298
          for hardware-outgoing; Sun, 23 Feb 1997 17:17:13 -0800 (PST)
Received: from gargoyle.bazzle.com (gargoyle.bazzle.com [206.103.246.190])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA15289
          for <hardware@freebsd.org>; Sun, 23 Feb 1997 17:17:05 -0800 (PST)
Received: from gargoyle.bazzle.com (net2.bazzle.com [206.103.246.189]) by gargoyle.bazzle.com (8.8.5/8.6.12) with SMTP id UAA09381; Sun, 23 Feb 1997 20:17:05 -0500 (EST)
Date: Sun, 23 Feb 1997 20:17:05 -0500 (EST)
From: "Eric J. Chet" <ejc@gargoyle.bazzle.com>
To: Ollivier Robert <roberto@keltia.freenix.fr>
cc: hardware@freebsd.org
Subject: Re: _big_ IDE disks?
In-Reply-To: <Pine.BSF.3.95.970223200907.9235B-100000@gargoyle.bazzle.com>
Message-ID: <Pine.BSF.3.95.970223201544.9365B-100000@gargoyle.bazzle.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Well
	I meant to say "much better in -current"

Eric

> FreeBSD 3.0-CURRENT #0: Fri Feb 21 22:59:34 EST 1997
>
> This is a GigaByte 586DX-dual pentium board.
> # dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 9.184039 secs (114173731 bytes/sec)
> 
> Eric 


From owner-freebsd-hardware  Mon Feb 24 00:10:28 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA07558
          for hardware-outgoing; Mon, 24 Feb 1997 00:10:28 -0800 (PST)
Received: from scruz.net (nic.scruz.net [165.227.1.2])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA07550
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 00:10:24 -0800 (PST)
Received: (from speth@localhost)
	by scruz.net (8.7.3/1.34) id AAA07385; Mon, 24 Feb 1997 00:10:10 -0800 (PST)
Message-Id: <199702240810.AAA07385@scruz.net>
From: speth@scruz.net (James G. Speth)
Date: Mon, 24 Feb 1997 00:10:09 -0800
In-Reply-To: Craig Shaver <craig@ProGroup.COM>
       "Re: problem with 2.1.7-RELEASE #0" (Feb 21, 21:51)
X-Mailer: Mail User's Shell (7.2.5 10/14/92)
To: Craig Shaver <craig@ProGroup.COM>
Subject: Re: problem with 2.1.7-RELEASE #0
Cc: hardware@freebsd.org, matthew@nic.scruz.net
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Craig Shaver wrote:
>
> I don't think it is anything wrong with FreeBSD 2.1.7.  I just built a
[del..]
> So, if you have some other memory available, try swapping it around.
> Try setting your bios to very conservative settings.  Disable the cache.

Yes, that was it exactly.  I slowed the clock down to 60 Mhz to try running
the RAM a bit slower and everything worked fine.  So I returned the RAM and
got some from a different manufacturer, and since then I've been running fine
at full speed.

Thanks,
Jim

From owner-freebsd-hardware  Mon Feb 24 00:19:49 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA07896
          for hardware-outgoing; Mon, 24 Feb 1997 00:19:49 -0800 (PST)
Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA07891
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 00:19:47 -0800 (PST)
Received: from opus.cts.cwu.edu (skynyrd@opus.cts.cwu.edu [198.104.92.71]) by pahtoh.cwu.edu (8.6.13/8.6.9) with ESMTP id AAA24000; Mon, 24 Feb 1997 00:19:46 -0800
Received: from localhost (skynyrd@localhost)
	by opus.cts.cwu.edu (8.8.5/8.8.5) with SMTP id AAA10404;
	Mon, 24 Feb 1997 00:19:45 -0800 (PST)
Date: Mon, 24 Feb 1997 00:19:44 -0800 (PST)
From: Chris Timmons <skynyrd@opus.cts.cwu.edu>
To: Nate Williams <nate@mt.sri.com>
cc: Bruce Evans <bde@zeta.org.au>, hardware@freebsd.org
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <199702231629.JAA11518@rocky.mt.sri.com>
Message-ID: <Pine.BSF.3.95.970224000851.9983G-100000@opus.cts.cwu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


I have a variety of systems, all with asus mb's, and running everything
from 2.1.7 up through current on p5's and p6's.  

The triton-1 machine running 2.1.7 consistently showed ~74MB/sec, while
the natoma p6-200 systems running ecc showed ~80MB/sec (both
just-before-lite2-current and 2.2 from about 1/26.)

the triton-2 machines (1 w/96MB edo, the other with 64MB fpm) both showed
rates ~117MB when the systems were otherwise unloaded.  These machines
both were running 2.2 from yesterday morning.

These results are with the standard bios parameters for 60ns memory.
There are a few memory knobs besides normal timing in the p6np5 bios but I
haven't experimented.  I suppose Rod Grimes would know what the optimal
settings are :)

-Chris


From owner-freebsd-hardware  Mon Feb 24 05:33:01 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id FAA19108
          for hardware-outgoing; Mon, 24 Feb 1997 05:33:01 -0800 (PST)
Received: from gvr.win.tue.nl (root@gvr.win.tue.nl [131.155.210.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA19102
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 05:32:58 -0800 (PST)
Received: (from guido@localhost) by gvr.win.tue.nl (8.8.5/8.8.2) id OAA02621 for freebsd-hardware@freebsd.org; Mon, 24 Feb 1997 14:32:54 +0100 (MET)
From: Guido van Rooij <guido@gvr.win.tue.nl>
Message-Id: <199702241332.OAA02621@gvr.win.tue.nl>
Subject: multi-interface ethernet cards
To: freebsd-hardware@freebsd.org
Date: Mon, 24 Feb 1997 14:32:54 +0100 (MET)
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Does any know if there are ethernet cards available (and supported) with
multiple interfaces (like 4).

-Guido

From owner-freebsd-hardware  Mon Feb 24 08:21:50 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA26715
          for hardware-outgoing; Mon, 24 Feb 1997 08:21:50 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA26704
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 08:21:42 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id DAA07615; Tue, 25 Feb 1997 03:18:17 +1100
Date: Tue, 25 Feb 1997 03:18:17 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702241618.DAA07615@godzilla.zeta.org.au>
To: nate@mt.sri.com, skynyrd@opus.cts.cwu.edu
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
Cc: bde@zeta.org.au, hardware@freebsd.org
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>The triton-1 machine running 2.1.7 consistently showed ~74MB/sec, while
>the natoma p6-200 systems running ecc showed ~80MB/sec (both
>just-before-lite2-current and 2.2 from about 1/26.)
>
>the triton-2 machines (1 w/96MB edo, the other with 64MB fpm) both showed
>rates ~117MB when the systems were otherwise unloaded.  These machines
>both were running 2.2 from yesterday morning.

Not good :-).  I have Triton-1 (ASUS P55TP4XE) with non-EDO RAM and the
benchmark runs at about 119.2MB/sec (1MB = 1048576).  Oops, I forgot
that I run with minor improvements that reduce the overhead of the inner
loop from 59 cycles to 57.  This probably accounts for the 119.2/117
difference.  Triton-2 and EDO RAM probably don't make much difference.

>These results are with the standard bios parameters for 60ns memory.
>There are a few memory knobs besides normal timing in the p6np5 bios but I
>haven't experimented.  I suppose Rod Grimes would know what the optimal
>settings are :)

I use the standard knobs with everything turned up high.  An x-2-2-2
write burst cycle is most important for this benchmark.

Bruce

Script started on Tue Feb 25 03:04:55 1997
ttyv1:bde@alphplex:/tmp> dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 8.389282 secs (124989958 bytes/sec)
ttyv1:bde@alphplex:/tmp> exit

Script done on Tue Feb 25 03:05:07 1997

Calibrating clock(s) relative to mc146818A clock ... i586 clock: 132622829 Hz, i8254 clock: 1193105 Hz
CPU: Pentium (132.62-MHz 586-class CPU)
  Origin = "GenuineIntel"  Id = 0x52b  Stepping=11
  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 33554432 (32768K bytes)
avail memory = 30412800 (29700K bytes)
DEVFS: ready for devices
Probing for devices on PCI bus 0:
chip0 <Intel 82437FX PCI cache memory controller> rev 2 on pci0:0:0

From owner-freebsd-hardware  Mon Feb 24 08:42:49 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA28112
          for hardware-outgoing; Mon, 24 Feb 1997 08:42:49 -0800 (PST)
Received: from seabass.progroup.com (catfish.progroup.com [206.24.122.2])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA28086
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 08:42:18 -0800 (PST)
Received: from seabass.progroup.com (seabass.progroup.com [206.24.122.1]) by seabass.progroup.com (8.7.5/8.7.3) with SMTP id IAA27183; Mon, 24 Feb 1997 08:39:43 -0800 (PST)
Message-ID: <3311C44F.5656AEC7@ProGroup.com>
Date: Mon, 24 Feb 1997 08:39:43 -0800
From: Craig Shaver <craig@ProGroup.com>
Organization: Productivity Group, Inc.
X-Mailer: Mozilla 3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386)
MIME-Version: 1.0
To: "James G. Speth" <speth@scruz.net>
CC: hardware@freebsd.org, matthew@nic.scruz.net
Subject: Re: problem with 2.1.7-RELEASE #0
References: <199702240810.AAA07385@scruz.net>
Content-Type: multipart/mixed; boundary="------------31D2DE924A7B7C1D61133CF4"
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

This is a multi-part message in MIME format.

--------------31D2DE924A7B7C1D61133CF4
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

James G. Speth wrote:
> 
> Craig Shaver wrote:
> >
> > I don't think it is anything wrong with FreeBSD 2.1.7.  I just built a
> [del..]
> > So, if you have some other memory available, try swapping it around.
> > Try setting your bios to very conservative settings.  Disable the cache.
> 
> Yes, that was it exactly.  I slowed the clock down to 60 Mhz to try running
> the RAM a bit slower and everything worked fine.  So I returned the RAM and
> got some from a different manufacturer, and since then I've been running fine
> at full speed.
> 
> Thanks,
> Jim

Outstanding, another satisfied customer!

I attached Tom's Hardware Page.  Check it out.  When I set up my Tyan
Tomcat III dual p5, I used the advice from this page to set up the
timings.  I purchased the p5-150's because that was the best bang for
the buck.  The first thing I did was push the bus speed from 60 to 66. 
And then I pushed the clock ratio from 2.5 to 3.  This gave me 200mhz
speeds.  Once I got the memory right, it has been running like a charm. 
I am keeping it running all the time to see if I can get it to overheat
and give me trouble.  So far the cpus are running cold.  I can touch
them without feeling any heat at all.  I am running Solaris 2.5.1 x86 on
it, and I throw a good load at it from time to time.  

The memory I had originally was from TI, and the replacement memory was
from Mu tech.

-- 
Craig Shaver  (craig@progroup.com) (415)390-0654 
Productivity Group POB 60458 Sunnyvale, CA  94088

--------------31D2DE924A7B7C1D61133CF4
Content-Type: text/html; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Base: "http://sysdoc.pair.com/"

<BASE HREF=3D"http://sysdoc.pair.com/">

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN">
<html>

<head>
<meta http-equiv=3D"Content-Type"
content=3D"text/html; charset=3Diso-8859-1">
<meta name=3D"Author" content=3D"Thomas Pabst">
<meta name=3D"GENERATOR" content=3D"Microsoft FrontPage 2.0">
<title>Tom's Hardware Guide</title>
</head>

<body background=3D"bgrnd.gif" bgcolor=3D"#FFFFFF" link=3D"#000000"
vlink=3D"#000000" alink=3D"#000000">
<!-- #$-:Image Map file created by Map THIS! --><!-- #$-:Map THIS! free i=
mage map editor by Todd C. Wilson --><!-- #$-:Please do not edit lines st=
arting with "#$" --><!-- #$VERSION:1.30 --><!-- #$AUTHOR:Thomas Pabst -->=
<!-- #$DATE:Sun Dec 29 03:22:12 1996 --><!-- #$PATH:G:\Guide\ --><!-- #$G=
IF:tomlogo1.gif -->
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"100%">
<tr>
<td><table border=3D"0" cellpadding=3D"0" cellspacing=3D"0"
width=3D"100%">
<tr>
<td width=3D"33%" bgcolor=3D"#00005A">&nbsp;</td>
<td width=3D"580"><!--webbot bot=3D"ImageMap"
startspan
rectangle=3D" (58,49) (144, 67)http://sysdoc.pair.com/"
rectangle=3D" (149,48) (242, 67)http://sysdoc.pair.com/news.html"
rectangle=3D" (247,47) (319, 66)http://sysdoc.pair.com/review.html"
rectangle=3D" (324,48) (393, 67)http://sysdoc.pair.com/surveys.html"
rectangle=3D" (398,48) (452, 67)http://sysdoc.pair.com/links.html"
rectangle=3D" (456,48) (525, 67)http://sysdoc.pair.com/contact.html"
rectangle=3D" (33,78) (128, 98)http://sysdoc.pair.com/mainboard.html"
rectangle=3D" (133,79) (182, 99)http://sysdoc.pair.com/cpu.html"
rectangle=3D" (187,79) (254, 98)http://sysdoc.pair.com/chipset.html"
rectangle=3D" (259,80) (310, 98)http://sysdoc.pair.com/ram.html"
rectangle=3D" (316,79) (389, 98)http://sysdoc.pair.com/hdd.html"
rectangle=3D" (394,79) (450, 98)http://sysdoc.pair.com/video.html"
rectangle=3D" (455,79) (550, 98)http://sysdoc.pair.com/performance.html"
src=3D"tomlogo1.gif" alt=3D"Navigation CSIM"
border=3D"0" width=3D"580" height=3D"105" --><MAP NAME=3D"FrontPageMap0">=
<AREA SHAPE=3D"RECT" COORDS=3D"58, 49, 144, 67" HREF=3D"http://sysdoc.pai=
r.com/"><AREA SHAPE=3D"RECT" COORDS=3D"149, 48, 242, 67" HREF=3D"http://s=
ysdoc.pair.com/news.html"><AREA SHAPE=3D"RECT" COORDS=3D"247, 47, 319, 66=
" HREF=3D"http://sysdoc.pair.com/review.html"><AREA SHAPE=3D"RECT" COORDS=
=3D"324, 48, 393, 67" HREF=3D"http://sysdoc.pair.com/surveys.html"><AREA =
SHAPE=3D"RECT" COORDS=3D"398, 48, 452, 67" HREF=3D"http://sysdoc.pair.com=
/links.html"><AREA SHAPE=3D"RECT" COORDS=3D"456, 48, 525, 67" HREF=3D"htt=
p://sysdoc.pair.com/contact.html"><AREA SHAPE=3D"RECT" COORDS=3D"33, 78, =
128, 98" HREF=3D"http://sysdoc.pair.com/mainboard.html"><AREA SHAPE=3D"RE=
CT" COORDS=3D"133, 79, 182, 99" HREF=3D"http://sysdoc.pair.com/cpu.html">=
<AREA SHAPE=3D"RECT" COORDS=3D"187, 79, 254, 98" HREF=3D"http://sysdoc.pa=
ir.com/chipset.html"><AREA SHAPE=3D"RECT" COORDS=3D"259, 80, 310, 98" HRE=
F=3D"http://sysdoc.pair.com/ram.html"><AREA SHAPE=3D"RECT" COORDS=3D"316,=
 79, 389, 98" HREF=3D"http://sysdoc.pair.com/hdd.html"><AREA SHAPE=3D"REC=
T" COORDS=3D"394, 79, 450, 98" HREF=3D"http://sysdoc.pair.com/video.html"=
><AREA SHAPE=3D"RECT" COORDS=3D"455, 79, 550, 98" HREF=3D"http://sysdoc.p=
air.com/performance.html"></MAP><img src=3D"tomlogo1.gif" alt=3D"Navigati=
on CSIM" border=3D"0" width=3D"580" height=3D"105" usemap=3D"#FrontPageMa=
p0"><!--webbot
bot=3D"ImageMap" i-checksum=3D"34395" endspan --></td>
<td width=3D"33%" bgcolor=3D"#0000F7">&nbsp;</td>
</tr>
</table>
</td>
</tr>
<tr>
<td><table border=3D"0" cellpadding=3D"2" width=3D"100%">
<tr>
<td valign=3D"top" rowspan=3D"2" width=3D"30%"><table
border=3D"0" cellpadding=3D"2" cellspacing=3D"0"
width=3D"100%">
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#00005A"><a href=3D"news.html"><img
src=3D"0whatsnew.gif" alt=3D"What's New"
border=3D"0" width=3D"91" height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#00005A"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"news.html"><font size=3D"2"
face=3D"Arial">last modified February 11, 1997</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#000064"><a href=3D"review.html"><img
src=3D"0rev.gif" border=3D"0" width=3D"71"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#000064"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"testlab.html"><font
size=3D"2" face=3D"Arial"><strong>The
Test Lab Page</strong></font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a href=3D"abitit5h.html"><font size=3D"2"
face=3D"Arial">ABIT IT5H</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a href=3D"abitit5v.html"><font size=3D"2"
face=3D"Arial">ABIT IT5V</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><font size=3D"2"
face=3D"Arial">ABIT PR5</font></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><font size=3D"2"
face=3D"Arial">ABIT PS6</font></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"asust2p4.html"><font
size=3D"2" face=3D"Arial">ASUS
P/I-P55T2P4 rev. 3</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">FIC PA-2005</font></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">FIC PA-2006</font></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">FIC VA-501</font></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">FKI SL586VT
II</font></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">Blitz'N
Software Cyborg</font></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#00006E"><a href=3D"surveys.html"><img
src=3D"0sur.gif" border=3D"0" width=3D"68"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#00006E"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"ctcm.html"><font size=3D"2"
face=3D"Arial">ctcm
Mainboard Benchmark</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"MBqueryform.html"><font
size=3D"2" face=3D"Arial">Mainboard
Survey</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"overclock.html#Toms Overclocking Survey"><font
size=3D"2" face=3D"Arial">Overclocking
Survey</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"75mhzbus.html"><font
size=3D"2" face=3D"Arial">75/83
MHz Bus Speed Survey</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"videobench.html"><font
size=3D"2" face=3D"Arial">Quake
1.06 Benchmark</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"mtm-survey.html"><font
size=3D"2" face=3D"Arial">Monster
Truck Madness Benchmark</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#000078"><a
href=3D"mainboard.html"><img src=3D"0mb.gif"
border=3D"0" width=3D"94" height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#000078"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"mainboard.html#whats important"><font
size=3D"2" face=3D"Arial">What
you should know</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"mainboard.html#results"><font
size=3D"2" face=3D"Arial">Test
Results</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"mainboard.html#recommendations"><font
size=3D"2" face=3D"Arial">Recommendations</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"mainboard.html#the future"><font
size=3D"2" face=3D"Arial">The
Future</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#000082"><a href=3D"cpu.html"><img
src=3D"0cpu.gif" border=3D"0" width=3D"48"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#000082"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"cpu.html#The Intel Pentium"><font
size=3D"2" face=3D"Arial"><b>The
Intel Pentium Processor</b></font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#MMX"><font
size=3D"2" face=3D"Arial">The
new MMX Technology</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#Things you should know"><font
size=3D"2" face=3D"Arial">Things
to be considered</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"pentium.html"><font
size=3D"2" face=3D"Arial">Pentium
Processor Reference Table</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"cpu.html#ppro"><font
size=3D"2" face=3D"Arial"><strong>The
Intel Pentium Pro</strong></font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#The Cyrix"><font
size=3D"2" face=3D"Arial"><b>The
Cyrix 6x86</b></font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#m2"><font size=3D"2"
face=3D"Arial">The next
Generation M2 6x86</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"cpu.html#686 links"><font
size=3D"2" face=3D"Arial">Hot
6x86 Links</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#tech 686"><font
size=3D"2" face=3D"Arial">Technical
Specifications of the 6x86</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#686 voltage"><font
size=3D"2" face=3D"Arial">The
6x86 Supply Voltage</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a
href=3D"cpu.html#Problems with the 6x86"><font
size=3D"2" face=3D"Arial">Problems
with the 6x86</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"cpu.html#AMD K5"><font
size=3D"2" face=3D"Arial"><b>The
AMD K5</b></font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#00008C"><a href=3D"chipset.html"><font
size=3D"2" face=3D"Arial"><img
src=3D"0chps.gif" border=3D"0" width=3D"65"
height=3D"17"></font></a></td>
</tr>
<tr>
<td bgcolor=3D"#00008C"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"chipset.html#intel-via"><font
size=3D"2" face=3D"Arial">Comparison
between the Intel and VIA
chipsets</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"triton1.html"><font
size=3D"2" face=3D"Arial">Specifications
of the Triton FX PCIset</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"http://www.intel.com/design/pcisets/prodbref/430hx_pb.htm"><font
size=3D"2" face=3D"Arial">Specifications
of the Triton HX PCIset</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"http://www.intel.com/design/pcisets/prodbref/430vx_pb.htm"><font
size=3D"2" face=3D"Arial">Specifications
of the Triton VX PCIset</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"vp1.html"><font size=3D"2"
face=3D"Arial">The VIA
Apollo VP-1 Chipset</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"vp2.html"><font size=3D"2"
face=3D"Arial">The VIA
Apollo VP-2 Chipset</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a
href=3D"http://www.intel.com/design/pcisets/prodbref/29772601.htm"><font
size=3D"2" face=3D"Arial">The
Intel 440FX Natoma PPro Chipset</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"VP680.html"><font
size=3D"2" face=3D"Arial">The
VIA Apollo P6 PPro Chipset</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#000096"><a href=3D"ram.html"><font
size=3D"2" face=3D"Arial"><img
src=3D"0ram.gif" border=3D"0" width=3D"50"
height=3D"17"></font></a></td>
</tr>
<tr>
<td bgcolor=3D"#000096"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"ram.html#SIMM/DIMM"><font
size=3D"2" face=3D"Arial">DIMMs
and SIMMs</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"ram.html#FPM"><font
size=3D"2" face=3D"Arial">FPM</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#EDO"><font
size=3D"2" face=3D"Arial">EDO</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#BEDO"><font
size=3D"2" face=3D"Arial">BEDO</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#SDRAM"><font
size=3D"2" face=3D"Arial">SDRAM</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#Async"><font
size=3D"2" face=3D"Arial">Async
SRAM</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#Sync"><font
size=3D"2" face=3D"Arial">Sync
SRAM</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#PB"><font size=3D"2"
face=3D"Arial">PB SRAM</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a
href=3D"ram.html#The Level2 Cache"><font
size=3D"2" face=3D"Arial">Level
2 Cache Size</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"ram.html#SRAM Recommendations"><font
size=3D"2" face=3D"Arial">SRAM
Recommendations</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000A0"><a href=3D"hdd.html"><font
size=3D"2" face=3D"Arial"><img
src=3D"0hdd.gif" border=3D"0" width=3D"73"
height=3D"17"></font></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000A0"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a
href=3D"hdd.html#EIDE FAQ"><font
size=3D"2" face=3D"Arial">EIDE/ATA
FAQ</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"hdd.html#Bus Master"><font
size=3D"2" face=3D"Arial">The
Bus Master DMA feature</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"hdd.html#DMA Driver"><font
size=3D"2" face=3D"Arial">Bus
Master DMA drivers</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"hdd.html#DMA Problems"><font
size=3D"2" face=3D"Arial">Troubleshooting
DMA drivers</font></a></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000AA"><a href=3D"video.html"><font
size=3D"2" face=3D"Arial"><img
src=3D"0vid.gif" border=3D"0" width=3D"55"
height=3D"17"></font></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000AA"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td width=3D"100%"><a
href=3D"video.html#The Basics"><font
size=3D"2" face=3D"Arial">The
Basics</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td><a href=3D"video.html#how work"><font
size=3D"2" face=3D"Arial">How
does it work ?</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><font size=3D"2"
face=3D"Arial">The Top 2D
Accelerators</font></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><a
href=3D"/cgi-bin/mtm-results.cgi"><font
size=3D"2" face=3D"Arial">The
Top 3D Accelerators</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td valign=3D"top"><font size=3D"2"
face=3D"Arial">Tom's
Ultimative Video Benchmark Survey</font></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000B4"><a
href=3D"performance.html"><img
src=3D"0perf.gif" border=3D"0" width=3D"94"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000B4"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top" width=3D"14"><font
size=3D"2" face=3D"Arial"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></font></td>
<td><a href=3D"overclock.html"><font
size=3D"2" face=3D"Arial"><strong>Overclocking
the CPU</strong></font></a><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0">
<tr>
<td valign=3D"top"><font
size=3D"2"
face=3D"Arial"><img
src=3D"blball.gif"
width=3D"14" height=3D"14"></font></td>
<td valign=3D"top"
width=3D"100%"><a
href=3D"overclock.html#OC requierements"><font
size=3D"2"
face=3D"Arial">Requirements</font></a></td>
</tr>
<tr>
<td valign=3D"top"><font
size=3D"2"
face=3D"Arial"><img
src=3D"blball.gif"
width=3D"14" height=3D"14"></font></td>
<td valign=3D"top"><a
href=3D"overclock.html#OC Theory"><font
size=3D"2"
face=3D"Arial">Theory</font></a></td>
</tr>
<tr>
<td valign=3D"top"><font
size=3D"2"
face=3D"Arial"><img
src=3D"blball.gif"
width=3D"14" height=3D"14"></font></td>
<td valign=3D"top"><a
href=3D"overclock.html#OC Practice"><font
size=3D"2"
face=3D"Arial">Practice</font></a></td>
</tr>
</table>
</td>
</tr>
<tr>
<td width=3D"14"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td><a href=3D"busspeed.html"><font
size=3D"2" face=3D"Arial"><strong>The
Importance of the Bus Speed</strong></font></a></td>
</tr>
<tr>
<td valign=3D"top" width=3D"14"><font
size=3D"2" face=3D"Arial"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></font></td>
<td><a href=3D"bios.html"><font
size=3D"2" face=3D"Arial"><strong>Configuring
the BIOS</strong></font></a><table
border=3D"0" cellpadding=3D"0"
cellspacing=3D"0">
<tr>
<td valign=3D"top"><font
size=3D"2"
face=3D"Arial"><img
src=3D"blball.gif"
width=3D"14" height=3D"14"></font></td>
<td valign=3D"top"
width=3D"100%"><a
href=3D"bios.html#memtiming"><font
size=3D"2"
face=3D"Arial">BIOS
Memory Timing</font></a></td>
</tr>
<tr>
<td valign=3D"top"><font
size=3D"2"
face=3D"Arial"><img
src=3D"blball.gif"
width=3D"14" height=3D"14"></font></td>
<td valign=3D"top"><a
href=3D"bios.html#links"><font
size=3D"2"
face=3D"Arial">BIOS
related Links</font></a></td>
</tr>
</table>
</td>
</tr>
<tr>
<td width=3D"14"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">Adjusting
the Chipset</font></td>
</tr>
<tr>
<td width=3D"14"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">Configuring
the HDD</font></td>
</tr>
<tr>
<td width=3D"14"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">Choosing
Drivers</font></td>
</tr>
<tr>
<td width=3D"14"><img
src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td><font size=3D"2"
face=3D"Arial">Choosing
Components</font></td>
</tr>
</table>
</center></div></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000BE"><a href=3D"links.html"><img
src=3D"0lin.gif" border=3D"0" width=3D"53"
height=3D"17"></a></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000C8"><a href=3D"contact.html"><img
src=3D"0con.gif" border=3D"0" width=3D"67"
height=3D"17"></a></td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000D2"><a href=3D"website.html"><img
src=3D"0webs.gif" border=3D"0" width=3D"95"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000D2">
<div align=3D"center"><center>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a href=3D"history.html"><font
size=3D"2" face=3D"Arial">The History</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"current.html"><font size=3D"2"
face=3D"Arial">The Current State</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"future.html"><font size=3D"2"
face=3D"Arial">The Future</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"awards.html"><font size=3D"2"
face=3D"Arial">Awards</font></a></td>
</tr>
<tr>
<td><img src=3D"blball.gif" width=3D"14"
height=3D"14"></td>
<td width=3D"100%" bgcolor=3D"#FFFFFF"><a href=3D"credits.html"><font
size=3D"2" face=3D"Arial">Credits</font></a></td>
</tr>
</table>
</center></div>
</td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000DC"><a href=3D"author.html"><img
src=3D"0auth.gif" border=3D"0" width=3D"86"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000DC">
<div align=3D"center"><center>
<table border=3D"0" cellpadding=3D"0" cellspacing=3D"0" width=3D"95%"
bgcolor=3D"#FFFFFF">
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top" width=3D"100%"><a href=3D"personal.html"><font
size=3D"2" face=3D"Arial">About the Author</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"roadrunner.html"><font size=3D"2"
face=3D"Arial">Tom's Roadrunner</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"dream.html"><font size=3D"2"
face=3D"Arial">Tom's Dream System</font></a></td>
</tr>
<tr>
<td valign=3D"top"><img src=3D"blball.gif"
width=3D"14" height=3D"14"></td>
<td valign=3D"top"><a href=3D"images.html"><font size=3D"2"
face=3D"Arial">Some Images</font></a></td>
</tr>
</table>
</center></div>
</td>
</tr>
<tr>
<td align=3D"center" width=3D"100%"
bgcolor=3D"#0000E6"><a
href=3D"sponsoring.html"><img
src=3D"0adv.gif" border=3D"0" width=3D"111"
height=3D"17"></a></td>
</tr>
<tr>
<td bgcolor=3D"#0000E6">&nbsp;</td>
</tr>
</table>
</td>
<td valign=3D"top" width=3D"70%"><table border=3D"0"
cellpadding=3D"0" cellspacing=3D"0" width=3D"100%">
<tr>
<td align=3D"center"><hr size=3D"2"
width=3D"50%">
<p><font size=3D"2" face=3D"Arial"><strong>In
cooperation with the excellent</strong></font><br>
<a href=3D"http://www.ix.de/ct/"><img
src=3D"ctmag.gif" border=3D"0" width=3D"257"
height=3D"29"></a></p>
<p align=3D"center"><a href=3D"http://sysdoc.pair.com/cgi-sys/cgiwrap/sys=
doc/sponsor.loc?index"><font
size=3D"2"><img src=3D"http://sysdoc.pair.com/cgi-sys/cgiwrap/sysdoc/spon=
sor.gif?index&856801986" border=3D"0" =

width=3D"468" height=3D"55"><br>
Please visit my sponsor</font></a>
</p>

<form
action=3D"http://sysdoc.pair.com/cgi-bin/pageselect.pl"
method=3D"POST">
<div align=3D"center"><center><table
border=3D"0">
<tr>
<td><p align=3D"center"><font
size=3D"4" face=3D"Arial"><b>Please select the
page you'd like to read: </b></font></p>
<p align=3D"center"><font
size=3D"4"><b><select
name=3D"pageselection" size=3D"1">
<option selected
value=3D"news.html">What's New at Tom's Site </option>
<option value=3D"klamath.html">The Intel Klamath</option>
<option value=3D"mmx.html">The Pentium MMX</option>
<option value=3D"review.html">Tom's Hardware Reviews </option>
<option value=3D"testlab.html">Tom's Hardware Test Lab </option>
<option value=3D"mainboard.html">The Mainboard Guide </option>
<option value=3D"ctcm.html">The Mainboard Benchmark Page </option>
<option value=3D"75mhzbus.html">The 75/83 MHz Bus Speed Project </option>=

<option value=3D"chipset.html">The Chipset Guide </option>
<option value=3D"cpu.html">The CPU Guide </option>
<option value=3D"overclock.html">The Overclocking Guide </option>
<option value=3D"ram.html">The RAM Guide </option>
<option value=3D"hdd.html">The Harddisk Guide </option>
<option value=3D"bios.html">The BIOS Guide </option>
<option value=3D"video.html">The Video Card Guide </option>
<option value=3D"videobench.html">Quake 1.06 Benchmark Page </option>
<option value=3D"mtm-survey.html">Monster Truck Madness Benchmark Page </=
option>
<option value=3D"performance.html">Performance Page </option>
<option value=3D"links.html">Useful Links </option>
<option value=3D"dream.html">Tom's Dream System </option>
<option value=3D"home.html">Tom's Homepage </option>
<option value=3D"personal.html">Who's Tom </option>
<option value=3D"sponsoring.html">Advertising Information </option>
</select></b></font>&nbsp;&nbsp;&nbsp;
<input type=3D"submit"
name=3D"selpage" value=3D"GO!"></p>
</td>
</tr>
</table>
</center></div>
</form>
<hr size=3D"2" width=3D"50%">
</td>
</tr>
<tr>
<td><table border=3D"0" cellpadding=3D"2"
width=3D"100%" bgcolor=3D"#FF0000" background=3D"0vred.jpg">
<tr>
<td align=3D"center" width=3D"100%"><img
src=3D"0anou.gif" width=3D"288"
height=3D"19" vspace=3D"4"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"4"
cellspacing=3D"4" width=3D"95%">
<tr>
<td bgcolor=3D"#FFD9D9"><font size=3D"4" face=3D"Arial"><strong>I</strong=
></font><font
size=3D"2" face=3D"Arial">n case you have been
reading my news page, you are aware of my huge project of
updating my entire website. I was planning to release all
the new pages as well as the altered current pages all at
once on one day. However the project is much too huge and
after all I am currently testing several new motherboards
as well. Hence I am releasing this renewed front page
today and with it the first new pages. There will be
quite a lot more soon and also the update of almost all
of my 'old' pages.</font>
<ul>
<li><img src=3D"hot.gif" width=3D"31" height=3D"12"><font size=3D"3" face=
=3D"Arial"><strong>First TX Board Review!! =

<a href=3D"aopenap5t.html">AOpen AP5T and AX5T</a>
 - How fast is Intel's new 430TX chipset ?</strong></font></li>
<li><img src=3D"hot.gif" width=3D"31" height=3D"12"><font size=3D"3" face=
=3D"Arial"><strong>
<a href=3D"klamath.html">The Intel Klamath Processor</a>
 - Pictures and Facts about the PPro Successor</strong></font></li>
<li><img src=3D"hot.gif" width=3D"31" height=3D"12"><font size=3D"3" face=
=3D"Arial"><strong>
<a href=3D"mmx.html">Does the Pentium MMX Live up to the Expectations?</a=
>
 Read my own experiences with the Pentium MMX !!!</strong></font></li>
<li><img src=3D"hot.gif" width=3D"31" height=3D"12"> <font size=3D"2" fac=
e=3D"Arial"><strong>
<a href=3D"performance.html">The Performance Guide</a> - after 10 months =
this page is finished!!!</strong></font></li>
<li><font size=3D"2" face=3D"Arial"><strong><img
src=3D"new2.gif" width=3D"31"
height=3D"12"> </strong></font><a
href=3D"hdd.html"><font size=3D"2"
face=3D"Arial"><strong>The Harddisk Guide</strong></font></a><font
size=3D"2" face=3D"Arial"><strong>, by
Frederik Gross, lots of top information about
harddisks and controllers</strong></font></li>
<li><font size=3D"2" face=3D"Arial"><strong><img
src=3D"new2.gif" width=3D"31"
height=3D"12"> </strong></font><a
href=3D"overclock.html"><font
size=3D"2" face=3D"Arial"><strong>The
Overclocking Guide</strong></font></a><font
size=3D"2" face=3D"Arial"><strong>, complete
new page, three times as big as before!!</strong></font></li>
</ul>
</td>
</tr>
<tr>
<td bgcolor=3D"#FFD9D9"><font
size=3D"4"
face=3D"Arial"><strong>T</strong></font><font
size=3D"2"
face=3D"Arial">here
are still quite a few
pages, which are of
particular interest to my
newer visitors.<br>
These are:</font><ul>
<li><a
href=3D"busspeed.html"><font
size=3D"2"
face=3D"Arial"><strong>The
Bus Speed Guide</strong></font></a></li>
<li><a
href=3D"75mhzbus.html"><font
size=3D"2"
face=3D"Arial"><strong>The
75/83 MHz Bus Speed
Project</strong></font></a></li>
<li><a href=3D"cpu.html"><font
size=3D"2"
face=3D"Arial"><strong>The
CPU Guide</strong></font></a></li>
<li><a
href=3D"mtm-survey.html"><font
size=3D"2"
face=3D"Arial"><strong>The
Monster Truck Madness
Direct3D Benchmark</strong></font></a></li>
<li><a
href=3D"videobench.html"><font
size=3D"2"
face=3D"Arial"><strong>The
Quake Timedemo Benchmark</strong></font></a></li>
</ul>
</td>
</tr>
<tr>
<td bgcolor=3D"#FFD9D9"><font
size=3D"4"
face=3D"Arial"><strong>Y</strong></font><font
size=3D"2"
face=3D"Arial">es,
and I am still on the
look for some more
sponsors!<br>
Advertising on this
website is the best a
hardware selling or
manufacturing company can
do.<br>
please visit my </font><a
href=3D"sponsoring.html"><font
size=3D"2"
face=3D"Arial"><strong>Advertising
Information Page</strong></font></a><font
size=3D"2"
face=3D"Arial"> for
further information.</font></td>
</tr>
</table>
</center></div></td>
</tr>
</table>
</td>
</tr>
<tr>
<td><hr size=3D"2" width=3D"50%">
<table border=3D"0" cellpadding=3D"2"
width=3D"100%" bgcolor=3D"#008000" background=3D"0vgreen.jpg">
<tr>
<td align=3D"center" width=3D"100%"><img
src=3D"0comm.gif" width=3D"185"
height=3D"19" vspace=3D"4"><div align=3D"center"><center><table
border=3D"0" cellpadding=3D"4"
cellspacing=3D"4" width=3D"95%">
<tr>
<td bgcolor=3D"#CEFFCE"><font
size=3D"4"
face=3D"Arial"><strong><b>T</b></strong></font><font
size=3D"2"
face=3D"Arial">his
Page is designed to help
and explain all the
important hardware
aspects of a modern
Pentium/Pentium Pro
machine. It should be
helpful to beginners as
well as to experienced
computer users to make
you able to understand
how your computer works
and how you can increase
its performance. </font></td>
</tr>
<tr>
<td bgcolor=3D"#CEFFCE"><font
size=3D"4"
face=3D"Arial"><strong>S</strong></font><font
size=3D"2"
face=3D"Arial">hould
there be any query
regarding the information
on this site, should you
have suggestions what I
could add, things you
always wanted to know and
ideas or complaints
regarding the
architecture or content
of my hardware pages then
feel free to tell me, ask
me or annoy me. </font><p
align=3D"center"><img
src=3D"smiley2.gif"
hspace=3D"20" width=3D"32"
height=3D"32"><a
href=3D"contact.html"><img
src=3D"email2.gif"
alt=3D"email me" border=3D"0"
width=3D"75" height=3D"50"></a><img
src=3D"smiley8.gif"
hspace=3D"20" width=3D"32"
height=3D"32"> </p>
</td>
</tr>
</table>
</center></div></td>
</tr>
</table>
</td>
</tr>
</table>
</td>
</tr>
<tr>
<td align=3D"right" valign=3D"bottom" width=3D"70%"><table
border=3D"0" cellpadding=3D"5" cellspacing=3D"5"
width=3D"100%">
<tr>
<td valign=3D"top" width=3D"50%"><hr size=3D"2"
width=3D"75%">
<p align=3D"center"><font size=3D"3"
face=3D"Arial"><strong>O</strong></font><font
size=3D"2" face=3D"Arial">uch ! This
was hit no.<strong> 1018577 </strong>since
September 21, 1996, 19:55 GMT</font></p>
<hr size=3D"2" width=3D"75%">
<p align=3D"center"><a
href=3D"http://www.pair.com/"><font
size=3D"2" face=3D"Arial"><img
src=3D"pair.gif" border=3D"0" width=3D"88"
height=3D"40"></font></a><font size=3D"2"
face=3D"Arial"><br>
First Class Website Hosting</font></p>
<hr size=3D"2" width=3D"75%">
<p align=3D"center"><font size=3D"2"
face=3D"Arial">Best experienced
with<br>
</font><a
href=3D"http://www.microsoft.com/ie/"><font
size=3D"2" face=3D"Arial"><img
src=3D"ie_static.gif" border=3D"0"
hspace=3D"10" vspace=3D"2" width=3D"88"
height=3D"31"></font></a><font size=3D"2"
face=3D"Arial"><br>
and I mean it !!</font></p>
<hr size=3D"2" width=3D"75%">
<p align=3D"center"><font size=3D"2" face=3D"Arial">
You are using Mozilla/3.01 (X11; I; FreeBSD 2.1.5-RELEASE i386)

</font></p>
<hr size=3D"2" width=3D"75%">
<p align=3D"center"><a
href=3D"http://www.microsoft.com/frontpage/"><font
size=3D"2" face=3D"Arial"><img
src=3D"fpeditor.gif"
alt=3D"Created with Microsoft FrontPage 97, although Front Page can hardl=
y handle this page ;-)"
border=3D"0" width=3D"16" height=3D"16"></font></a></p>
</td>
<td align=3D"center" valign=3D"top"
width=3D"50%" bgcolor=3D"#000000"><font
size=3D"2" face=3D"Arial"><img
src=3D"tab-legal.gif" width=3D"93"
height=3D"13"></font><table border=3D"0"
cellpadding=3D"5" cellspacing=3D"5"
width=3D"100%">
<tr>
<td valign=3D"top" width=3D"50%"
bgcolor=3D"#FFFFFF"><font size=3D"3"
face=3D"Arial"><strong>C</strong></font><font
size=3D"2" face=3D"Arial">opyright
of all documents and scripts
belonging to this site by </font><a
href=3D"mailto:tom@sysdoc.pair.com"><font
size=3D"2" face=3D"Arial">Thomas
Pabst</font></a><font size=3D"2"
face=3D"Arial"> 1996 - 1997.</font><p><font
size=3D"3" face=3D"Arial"><strong>M</strong></font><font
size=3D"2" face=3D"Arial">ost
of the information contained on
this site is copyrighted
material. It is illegal to copy
or redistribute this information
in any way without the expressed
written consent of the author.</font></p>
<p><font size=3D"3"
face=3D"Arial"><strong>T</strong></font><font
size=3D"2" face=3D"Arial">his
site is NOT responsible for any
damage that the information on
this site may cause to your
system. </font></p>
</td>
</tr>
</table>
</td>
</tr>
</table>
</td>
</tr>
</table>
</td>
</tr>
</table>
</body>
</html>

--------------31D2DE924A7B7C1D61133CF4--


From owner-freebsd-hardware  Mon Feb 24 08:47:38 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA28441
          for hardware-outgoing; Mon, 24 Feb 1997 08:47:38 -0800 (PST)
Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA28435
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 08:47:34 -0800 (PST)
Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA15355; Mon, 24 Feb 1997 09:47:26 -0700 (MST)
Date: Mon, 24 Feb 1997 09:47:26 -0700 (MST)
Message-Id: <199702241647.JAA15355@rocky.mt.sri.com>
From: Nate Williams <nate@mt.sri.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Bruce Evans <bde@zeta.org.au>
Cc: nate@mt.sri.com, skynyrd@opus.cts.cwu.edu, hardware@freebsd.org
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <199702241618.DAA07615@godzilla.zeta.org.au>
References: <199702241618.DAA07615@godzilla.zeta.org.au>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> >The triton-1 machine running 2.1.7 consistently showed ~74MB/sec, while
> >the natoma p6-200 systems running ecc showed ~80MB/sec (both
> >just-before-lite2-current and 2.2 from about 1/26.)
> >
> >the triton-2 machines (1 w/96MB edo, the other with 64MB fpm) both showed
> >rates ~117MB when the systems were otherwise unloaded.  These machines
> >both were running 2.2 from yesterday morning.
> 
> Not good :-).  I have Triton-1 (ASUS P55TP4XE) with non-EDO RAM and the
> benchmark runs at about 119.2MB/sec (1MB = 1048576).

Hmm, that's my motherboard, so I should be seeing similar results.

Which BIOS are you using, and which revision of the motherboard?

> >These results are with the standard bios parameters for 60ns memory.
> >There are a few memory knobs besides normal timing in the p6np5 bios but I
> >haven't experimented.  I suppose Rod Grimes would know what the optimal
> >settings are :)
> 
> I use the standard knobs with everything turned up high.  An x-2-2-2
> write burst cycle is most important for this benchmark.

I'm not sure what mine is set to.   I'll have to go check, and get back
to you on it.



Nate

From owner-freebsd-hardware  Mon Feb 24 08:54:29 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA29023
          for hardware-outgoing; Mon, 24 Feb 1997 08:54:29 -0800 (PST)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA29003
          for <hardware@FreeBSD.ORG>; Mon, 24 Feb 1997 08:54:23 -0800 (PST)
Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) 
  by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id IAA04704;
  Mon, 24 Feb 1997 08:52:08 -0800 (PST)
Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997))  id 88256448.005C4D89 ; Mon, 24 Feb 1997 08:48:11 -0700
X-Lotus-FromDomain: IW @ IWP
From: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
To: bde@zeta.org.au, nate@mt.sri.com, skynyrd@opus.cts.cwu.edu
cc: hardware@FreeBSD.ORG
Message-ID: <88256448.005C3197.00@IWND1.infoworld.com>
Date: Mon, 24 Feb 1997 09:45:00 -0700
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
Mime-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk






With what disk and controller are you getting 120 MB/sec transfer rates?



From owner-freebsd-hardware  Mon Feb 24 08:57:14 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id IAA29193
          for hardware-outgoing; Mon, 24 Feb 1997 08:57:14 -0800 (PST)
Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA29186
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 08:57:10 -0800 (PST)
Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA15409; Mon, 24 Feb 1997 09:57:01 -0700 (MST)
Date: Mon, 24 Feb 1997 09:57:01 -0700 (MST)
Message-Id: <199702241657.JAA15409@rocky.mt.sri.com>
From: Nate Williams <nate@mt.sri.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
Cc: hardware@freebsd.org
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <88256448.005C3197.00@IWND1.infoworld.com>
References: <88256448.005C3197.00@IWND1.infoworld.com>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> With what disk and controller are you getting 120 MB/sec transfer rates?
> 

Umm, read the articles in question.  It explains how we're doing memory
speed tests, not disk speed tests.  The disks aren't even being touched
(except to load the test program).


Nate

From owner-freebsd-hardware  Mon Feb 24 09:02:58 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id JAA29531
          for hardware-outgoing; Mon, 24 Feb 1997 09:02:58 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA29524
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 09:02:54 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id DAA08346; Tue, 25 Feb 1997 03:57:15 +1100
Date: Tue, 25 Feb 1997 03:57:15 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702241657.DAA08346@godzilla.zeta.org.au>
To: bde@zeta.org.au, nate@mt.sri.com
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
Cc: hardware@freebsd.org, skynyrd@opus.cts.cwu.edu
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>> Not good :-).  I have Triton-1 (ASUS P55TP4XE) with non-EDO RAM and the
>> benchmark runs at about 119.2MB/sec (1MB = 1048576).

>Which BIOS are you using, and which revision of the motherboard?

Award BIOS from about 1/1/96 and 2.4 motherboard.

Bruce

From owner-freebsd-hardware  Mon Feb 24 10:19:26 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id KAA05203
          for hardware-outgoing; Mon, 24 Feb 1997 10:19:26 -0800 (PST)
Received: from pahtoh.cwu.edu (root@pahtoh.cwu.edu [198.104.65.27])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA05186
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 10:19:23 -0800 (PST)
Received: from opus.cts.cwu.edu (skynyrd@opus.cts.cwu.edu [198.104.92.71]) by pahtoh.cwu.edu (8.6.13/8.6.9) with ESMTP id KAA26408; Mon, 24 Feb 1997 10:19:21 -0800
Received: from localhost (skynyrd@localhost)
	by opus.cts.cwu.edu (8.8.5/8.8.5) with SMTP id KAA12371;
	Mon, 24 Feb 1997 10:19:16 -0800 (PST)
Date: Mon, 24 Feb 1997 10:19:16 -0800 (PST)
From: Chris Timmons <skynyrd@opus.cts.cwu.edu>
To: Bruce Evans <bde@zeta.org.au>
cc: nate@mt.sri.com, hardware@freebsd.org
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <199702241618.DAA07615@godzilla.zeta.org.au>
Message-ID: <Pine.BSF.3.95.970224101131.11756D-100000@opus.cts.cwu.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


Hmmm....

On the p/i p6np5 pro200 machine (award bios) permits x-2-2-3 as the most
aggressive timing for both read/write (which were set.)

There is also "RAS Precharge" and "RAS to CAS" delay which are unexplained
in the manual.  Their values "3T" each by default seem to be a divisor, as
when I lowered the delay to 2T, I immediately started getting the kinds of
results in the benchmark that the other fellow was talking about on his
ppro - ~65MB.

The RAS Precharge could go to 4T, which I set, and realized a slight
performance gain in the benchmark to ~85MB, up from ~80MB.  I'll be
interested to see how this affects stability and if there is a measurable
gain on the make world times (-current building yet? :)

Thanks for the info about this... I'll fiddle with the t2p4 boards later
on.

-Chris



On Tue, 25 Feb 1997, Bruce Evans wrote:

> I use the standard knobs with everything turned up high.  An x-2-2-2
> write burst cycle is most important for this benchmark.
> 
> Bruce
> 
> Script started on Tue Feb 25 03:04:55 1997
> ttyv1:bde@alphplex:/tmp> dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 8.389282 secs (124989958 bytes/sec)
> ttyv1:bde@alphplex:/tmp> exit
> 
> Script done on Tue Feb 25 03:05:07 1997
> 
> Calibrating clock(s) relative to mc146818A clock ... i586 clock: 132622829 Hz, i8254 clock: 1193105 Hz
> CPU: Pentium (132.62-MHz 586-class CPU)
>   Origin = "GenuineIntel"  Id = 0x52b  Stepping=11
>   Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
> real memory  = 33554432 (32768K bytes)
> avail memory = 30412800 (29700K bytes)
> DEVFS: ready for devices
> Probing for devices on PCI bus 0:
> chip0 <Intel 82437FX PCI cache memory controller> rev 2 on pci0:0:0
> 


From owner-freebsd-hardware  Mon Feb 24 10:20:26 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id KAA05384
          for hardware-outgoing; Mon, 24 Feb 1997 10:20:26 -0800 (PST)
Received: from mail.ruhrgebiet.individual.net (in-ruhr.ruhr.de [141.39.224.38])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA05377
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 10:20:23 -0800 (PST)
Received: from robkaos.ruhr.de (admin@localhost)
	by mail.ruhrgebiet.individual.net (8.8.5/8.8.5) with UUCP id TAA13750
	for freebsd.org!freebsd-hardware; Mon, 24 Feb 1997 19:07:16 +0100 (MET)
Received: by robkaos.ruhr.de (/\oo/\ Smail3.1.29.1 #29.1)
	id <m0vz4kU-00065ZC@robkaos.ruhr.de>; Mon, 24 Feb 97 19:03 MET
Message-Id: <m0vz4kU-00065ZC@robkaos.ruhr.de>
From: robsch@robkaos.ruhr.de (Robert Schien)
Subject: Memory speed of P6-200 (256k)
To: freebsd-hardware@freebsd.org
Date: Mon, 24 Feb 1997 19:03:38 +0100 (MET)
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

I found:
dd if=/dev/zero of=/dev/null bs=1m count=2000 
2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
The kernel is 3.0-current.

Is this value normal for a P6-200?
If not, how can I speed it up?

TIA
Robert

From owner-freebsd-hardware  Mon Feb 24 10:53:07 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id KAA07826
          for hardware-outgoing; Mon, 24 Feb 1997 10:53:07 -0800 (PST)
Received: from persprog.com (persprog.com [204.215.255.203])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA07819
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 10:53:03 -0800 (PST)
Received:  by persprog.com (8.7.5/4.10)
	id NAA18845; Mon, 24 Feb 1997 13:49:01 -0500
Message-Id: <199702241849.NAA18845@persprog.com>
Received: from dasa(192.2.2.199) by cerberus.ppi.com via smap (V1.3)
	id sma018841; Mon Feb 24 13:48:47 1997
Received: from DASA/SpoolDir by dasa.ppi.com (Mercury 1.21);
    24 Feb 97 13:48:49 +0500
Received: from SpoolDir by DASA (Mercury 1.30); 24 Feb 97 13:48:37 +0500
From: "David Alderman" <dave@persprog.com>
Organization: Personalized Programming, Inc
To: Craig Shaver <craig@ProGroup.com>
Date: Mon, 24 Feb 1997 13:48:35 +0500
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Re: problem with 2.1.7-RELEASE #0
CC: hardware@freebsd.org
Priority: normal
In-reply-to: <3311C44F.5656AEC7@ProGroup.com>
X-mailer: Pegasus Mail for Win32 (v2.52)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On 24 Feb 97 at 8:39, Craig Shaver proclaimed:

> The memory I had originally was from TI, and the replacement memory was
> from Mu tech.
> 

Don't you mean Micron Technology (or were you just being humorous)?  
Micron uses a lower-case "mu" on their RAM.

Hey, maybe we should make up some other names for companies based on 
their logos.   I have always referred to United Microelectronic 
Corp as "Masters of the Universe" because their symbol is a globe 
with the M and U making up the circumference.   General Electric's 
symbol is often referred to as "the meatball".  AT&T is "the 
deathstar".   

I hadn't heard Mu Tech before.

Sorry everyone.  It's been a bad Monday and I think I'm suffering 
from sleep deprivation...




======================================
When philosophy conflicts with reality, choose fantasy.
Dave Alderman  -- dave@persprog.com
======================================

From owner-freebsd-hardware  Mon Feb 24 11:04:33 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA08620
          for hardware-outgoing; Mon, 24 Feb 1997 11:04:33 -0800 (PST)
Received: from Campino.Informatik.RWTH-Aachen.DE (campino.Informatik.RWTH-Aachen.DE [137.226.116.240])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA08594
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 11:04:21 -0800 (PST)
Received: from gil.physik.rwth-aachen.de (gilberto.physik.rwth-aachen.de [137.226.31.2]) by Campino.Informatik.RWTH-Aachen.DE (RBI-Z-5/8.6.12)
	with ESMTP id UAA03756; Mon, 24 Feb 1997 20:04:20 +0100 (MET)
Received: (from kuku@localhost) by gil.physik.rwth-aachen.de (8.8.4/8.6.9) id UAA01301; Mon, 24 Feb 1997 20:09:20 +0100 (MET)
From: Christoph Kukulies <kuku@gilberto.physik.RWTH-Aachen.DE>
Message-Id: <199702241909.UAA01301@gil.physik.rwth-aachen.de>
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <m0vz4kU-00065ZC@robkaos.ruhr.de> from Robert Schien at "Feb 24, 97 07:03:38 pm"
To: robsch@robkaos.ruhr.de (Robert Schien)
Date: Mon, 24 Feb 1997 20:09:20 +0100 (MET)
Cc: freebsd-hardware@freebsd.org
Reply-To: Christoph Kukulies <kuku@gilberto.physik.RWTH-Aachen.DE>
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> I found:
> dd if=/dev/zero of=/dev/null bs=1m count=2000 
> 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
> The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
> The kernel is 3.0-current.

This is from my P6NP5:

bach> dd if=/dev/zero of=/dev/null bs=1m count=2000
2000+0 records in
2000+0 records out
2097152000 bytes transferred in 25.741665 secs (81469167 bytes/sec)      

> 
> Is this value normal for a P6-200?
> If not, how can I speed it up?
> 
> TIA
> Robert
> 

--
Chris Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de

From owner-freebsd-hardware  Mon Feb 24 11:32:13 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA10393
          for hardware-outgoing; Mon, 24 Feb 1997 11:32:13 -0800 (PST)
Received: from george.lbl.gov (george.lbl.gov [128.3.196.93])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA10388
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 11:32:11 -0800 (PST)
Received: (jin@localhost) by george.lbl.gov (8.6.10/8.6.5) id LAA24844; Mon, 24 Feb 1997 11:27:26 -0800
Date: Mon, 24 Feb 1997 11:27:26 -0800
From: "Jin Guojun[ITG]" <jin@george.lbl.gov>
Message-Id: <199702241927.LAA24844@george.lbl.gov>
To: freebsd-hardware@freebsd.org, robsch@robkaos.ruhr.de
Subject: Re: Memory speed of P6-200 (256k)
Cc: skynyrd@opus.cts.cwu.edu
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Message-1:
} >> Not good :-).  I have Triton-1 (ASUS P55TP4XE) with non-EDO RAM and the
} >> benchmark runs at about 119.2MB/sec (1MB = 1048576).
} 
} >Which BIOS are you using, and which revision of the motherboard?
} 
} Award BIOS from about 1/1/96 and 2.4 motherboard.
} 
} Bruce

Message-2:
} I found:
} dd if=/dev/zero of=/dev/null bs=1m count=2000 
} 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
} The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
} The kernel is 3.0-current.
} 
} Is this value normal for a P6-200?
} If not, how can I speed it up?
} 
} TIA
} Robert

I do not know what is the orginal messurement, but it looks very interesting.
By invesgating such idea, I found that 2.1.x and 2.2 or higher have different
memory sub-system or dd implementation.

So, you guys should not worry any memory speed on your system, and there is
no thing can change the speed beside change the memory sub-system, such as
inteleaved memory system.

The command "dd if=/dev/zero of=/dev/null bs=1m count=XXX"

really does is to allocate 1MB memory and try to put zero into the piece
of memory and repeat XXX times.

Under 2.1.x-RELEASE, it looks like dd/memory-sub-system (more like dd)
uses 4-bytes aligned register to memory operation, and 2.2/higher use
8-bytes aligned register to memory operation.
I tested this operation under 6-different mortherboard and 4-different
CPU speed.  I have not compare the dd or memory sub-system between 2.1.x and
2.2/higher yet; but I am sure I got the right assumption. 

So, for these people want to know what is your best memory system performance,
refer to ftp://george.lbl.gov/pub/ccs/performance.ps, page 6-8, and look for
register to memory copy and memory copy.

Do be comfused by this dd command :-)

-Jin


From owner-freebsd-hardware  Mon Feb 24 11:35:27 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA10634
          for hardware-outgoing; Mon, 24 Feb 1997 11:35:27 -0800 (PST)
Received: from mexico.brainstorm.eu.org (root@mexico.brainstorm.fr [193.56.58.253])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA10626
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 11:35:21 -0800 (PST)
Received: from brasil.brainstorm.eu.org (brasil.brainstorm.fr [193.56.58.33])
          by mexico.brainstorm.eu.org (8.8.4/8.8.4) with ESMTP
	  id UAA04937 for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 20:35:06 +0100
Received: (from uucp@localhost) by brasil.brainstorm.eu.org (8.8.4/8.6.12) with UUCP id UAA01883 for freebsd-hardware@freebsd.org; Mon, 24 Feb 1997 20:35:00 +0100
Received: (from roberto@localhost) by keltia.freenix.fr (8.8.5/keltia-uucp-2.9) id UAA04246;
          Mon, 24 Feb 1997 20:17:01 +0100 (CET)
Message-ID: <19970224201700.PF29028@keltia.freenix.fr>
Date: Mon, 24 Feb 1997 20:17:00 +0100
From: roberto@keltia.freenix.fr (Ollivier Robert)
To: freebsd-hardware@freebsd.org
Subject: Re: multi-interface ethernet cards
References: <199702241332.OAA02621@gvr.win.tue.nl>
X-Mailer: Mutt 0.60,1-3,9
Mime-Version: 1.0
X-Operating-System: FreeBSD 3.0-CURRENT ctm#2999
In-Reply-To: <199702241332.OAA02621@gvr.win.tue.nl>; from Guido van Rooij on Feb 24, 1997 14:32:54 +0100
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

According to Guido van Rooij:
> Does any know if there are ethernet cards available (and supported) with
> multiple interfaces (like 4).

The Znyx cards have 4 ethernets and as they use the DEC chipset, we support
them.
-- 
Ollivier ROBERT    -=- The daemon is FREE! -=-    roberto@keltia.freenix.fr
  FreeBSD keltia.freenix.fr 3.0-CURRENT #39: Sun Feb  2 22:12:44 CET 1997

From owner-freebsd-hardware  Mon Feb 24 11:35:30 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA10650
          for hardware-outgoing; Mon, 24 Feb 1997 11:35:30 -0800 (PST)
Received: from mexico.brainstorm.eu.org (root@mexico.brainstorm.fr [193.56.58.253])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA10627
          for <hardware@freebsd.org>; Mon, 24 Feb 1997 11:35:21 -0800 (PST)
Received: from brasil.brainstorm.eu.org (brasil.brainstorm.fr [193.56.58.33])
          by mexico.brainstorm.eu.org (8.8.4/8.8.4) with ESMTP
	  id UAA04943; Mon, 24 Feb 1997 20:35:08 +0100
Received: (from uucp@localhost) by brasil.brainstorm.eu.org (8.8.4/8.6.12) with UUCP id UAA01882; Mon, 24 Feb 1997 20:34:59 +0100
Received: (from roberto@localhost) by keltia.freenix.fr (8.8.5/keltia-uucp-2.9) id UAA04229;
          Mon, 24 Feb 1997 20:14:52 +0100 (CET)
Message-ID: <19970224201451.HL52176@keltia.freenix.fr>
Date: Mon, 24 Feb 1997 20:14:51 +0100
From: roberto@keltia.freenix.fr (Ollivier Robert)
To: craig@ProGroup.com (Craig Shaver)
Cc: hardware@freebsd.org (Hardware Mailing list)
Subject: Re: problem with 2.1.7-RELEASE #0
References: <199702240810.AAA07385@scruz.net> <3311C44F.5656AEC7@ProGroup.com>
X-Mailer: Mutt 0.60,1-3,9
Mime-Version: 1.0
X-Operating-System: FreeBSD 3.0-CURRENT ctm#2999
In-Reply-To: <3311C44F.5656AEC7@ProGroup.com>; from Craig Shaver on Feb 24, 1997 08:39:43 -0800
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

According to Craig Shaver:
> I attached Tom's Hardware Page.  Check it out.  When I set up my Tyan
> Tomcat III dual p5, I used the advice from this page to set up the

Please *NEVER* do that again. Some of us actually pay for every byte
transmitted though our modem.
-- 
Ollivier ROBERT    -=- The daemon is FREE! -=-    roberto@keltia.freenix.fr
  FreeBSD keltia.freenix.fr 3.0-CURRENT #39: Sun Feb  2 22:12:44 CET 1997

From owner-freebsd-hardware  Mon Feb 24 11:42:35 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA11168
          for hardware-outgoing; Mon, 24 Feb 1997 11:42:35 -0800 (PST)
Received: from george.lbl.gov (george.lbl.gov [128.3.196.93])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA11158
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 11:42:32 -0800 (PST)
Received: (jin@localhost) by george.lbl.gov (8.6.10/8.6.5) id LAA25169 for freebsd-hardware@freebsd.org; Mon, 24 Feb 1997 11:37:49 -0800
Date: Mon, 24 Feb 1997 11:37:49 -0800
From: "Jin Guojun[ITG]" <jin@george.lbl.gov>
Message-Id: <199702241937.LAA25169@george.lbl.gov>
To: freebsd-hardware@freebsd.org
Subject: Re: multi-interface ethernet cards
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> According to Guido van Rooij:
> > Does any know if there are ethernet cards available (and supported) with
> > multiple interfaces (like 4).
> 
> The Znyx cards have 4 ethernets and as they use the DEC chipset, we support
> them.

SMC has 2-port fast ethernet card using DEC 21140 chipset, but a few vendors
sell it.

-Jin


From owner-freebsd-hardware  Mon Feb 24 15:35:52 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id PAA29622
          for hardware-outgoing; Mon, 24 Feb 1997 15:35:52 -0800 (PST)
Received: from pegasus.com (pegasus.com [140.174.243.13])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA29614
          for <freebsd-hardware@FreeBSD.org>; Mon, 24 Feb 1997 15:35:40 -0800 (PST)
Received: by pegasus.com (8.6.8/PEGASUS-2.2)
	id NAA22803; Mon, 24 Feb 1997 13:35:06 -1000
Date: Mon, 24 Feb 1997 13:35:06 -1000
From: richard@pegasus.com (Richard Foulk)
Message-Id: <199702242335.NAA22803@pegasus.com>
In-Reply-To: roberto@keltia.freenix.fr (Ollivier Robert)
       "Re: multi-interface ethernet cards" (Feb 24,  8:17pm)
X-Mailer: Mail User's Shell (7.2.5 10/14/92)
To: roberto@keltia.freenix.fr (Ollivier Robert), freebsd-hardware@FreeBSD.org
Subject: Re: multi-interface ethernet cards
Sender: owner-hardware@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

} > Does any know if there are ethernet cards available (and supported) with
} > multiple interfaces (like 4).
} 
} The Znyx cards have 4 ethernets and as they use the DEC chipset, we support
} them.
} -- 

What are the prices like on these cards and where can you order them from?


Thanks

Richard

From owner-freebsd-hardware  Mon Feb 24 17:03:30 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id RAA06328
          for hardware-outgoing; Mon, 24 Feb 1997 17:03:30 -0800 (PST)
Received: from mail0.iij.ad.jp (root@mail0.iij.ad.jp [192.244.176.61])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA06323
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 17:03:25 -0800 (PST)
Received: from uucp2.iij.ad.jp (uucp2.iij.ad.jp [192.244.176.74]) by mail0.iij.ad.jp (8.6.12+2.4W/3.3W9-MAIL) with ESMTP id KAA07243 for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 10:03:20 +0900
Received: from akagi.UUCP (uucp@localhost) by uucp2.iij.ad.jp (8.6.12+2.4W/3.3W9-UUCP) with UUCP id KAA03918 for freebsd-hardware@freebsd.org; Tue, 25 Feb 1997 10:03:19 +0900
Received: by akagi.ni-ait.co.jp (4.1/SMI-4.1)
	id AA07700; Tue, 25 Feb 97 10:04:10 JST
From: lab@ni-ait.co.jp (Larry Baird)
Message-Id: <9702250104.AA07700@akagi.ni-ait.co.jp>
Subject: Re: multi-interface ethernet cards
To: freebsd-hardware@freebsd.org
Date: Tue, 25 Feb 1997 10:04:10 +0900 (JST)
In-Reply-To: <199702242019.PAA16745@gta.gta.com> from "Jin Guojun[ITG]" at Feb 24, 97 11:37:49 am
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> > According to Guido van Rooij:
> > > Does any know if there are ethernet cards available (and supported) with
> > > multiple interfaces (like 4).
> > 
> > The Znyx cards have 4 ethernets and as they use the DEC chipset, we support
> > them.
Beware of the 100MB 4 port Zynx cards.  I have verified that the 100MB version
does not work with the driver for 2.1.6.  Has the driver for current been
changed to support the newer DEC chip sets?  Matt Thomas's has a driver
that is a drop in replacement for the current driver.  It seems to support
more cards that the standard but the last I checked didn't support the 100MB
4 port Zynx card.  Though I know he was working on getting such a card.

-- 
Larry Baird
Global Technology Associates, Inc.
3504 Lake Lynda Drive, Suite 160
Orlando, FL 32817
http://www.gta.com/
Tel: +1.407.380.0220
Fax: +1.407.380.6080

From owner-freebsd-hardware  Mon Feb 24 18:02:00 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id SAA10083
          for hardware-outgoing; Mon, 24 Feb 1997 18:02:00 -0800 (PST)
Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA10078
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 18:01:57 -0800 (PST)
Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.7.5/8.6.12) with SMTP id VAA06415; Mon, 24 Feb 1997 21:02:35 -0500 (EST)
Date: Mon, 24 Feb 1997 21:02:35 -0500 (EST)
From: Mark Mayo <mark@quickweb.com>
To: Christoph Kukulies <kuku@gilberto.physik.rwth-aachen.de>
cc: Robert Schien <robsch@robkaos.ruhr.de>, freebsd-hardware@freebsd.org
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <199702241909.UAA01301@gil.physik.rwth-aachen.de>
Message-ID: <Pine.BSF.3.95q.970224210037.6213D-100000@vinyl.quickweb.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

On Mon, 24 Feb 1997, Christoph Kukulies wrote:

> > I found:
> > dd if=/dev/zero of=/dev/null bs=1m count=2000 
> > 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
> > The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
> > The kernel is 3.0-current.
> 
> This is from my P6NP5:
> 
> bach> dd if=/dev/zero of=/dev/null bs=1m count=2000
> 2000+0 records in
> 2000+0 records out
> 2097152000 bytes transferred in 25.741665 secs (81469167 bytes/sec)      

While we're at it, here are the results fom my Intel 440FX(Natoma) P6-200
machine, running 2.1.5:

mark:{27}/home/mark % dd if=/dev/zero of=/dev/null bs=1m count=2000
2000+0 records in
2000+0 records out
2097152000 bytes transferred in 24 secs (87381333 bytes/sec)

My first generation PPro 150 from Digital only musters ~60MB/s..

-Mark

> 
> > 
> > Is this value normal for a P6-200?
> > If not, how can I speed it up?
> > 
> > TIA
> > Robert
> > 
> 
> --
> Chris Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de
> 

----------------------------------------------------------------------------
 Mark Mayo		  				mark@quickweb.com       
 RingZero Comp.  	  		   http://vinyl.quickweb.com/mark 
----------------------------------------------------------------------------
 Nature shows that with the growth of intelligence comes increased capacity 
 for pain, and it is only with the highest degree of intelligence that 
 suffering reaches its supreme point.  -- Arthur Schopenhauer


From owner-freebsd-hardware  Mon Feb 24 19:17:58 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id TAA15981
          for hardware-outgoing; Mon, 24 Feb 1997 19:17:58 -0800 (PST)
Received: from dfw-ix6.ix.netcom.com (dfw-ix6.ix.netcom.com [206.214.98.6])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA15958
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 19:17:54 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix6.ix.netcom.com (8.8.4/8.8.4)
	  id VAA04669; Mon, 24 Feb 1997 21:16:08 -0600 (CST)
Received: from ala-ca13-23.ix.netcom.com(204.32.168.55) by dfw-ix6.ix.netcom.com via smap (V1.3)
	id sma004493; Mon Feb 24 21:15:23 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id TAA02667; Mon, 24 Feb 1997 19:12:59 -0800 (PST)
Date: Mon, 24 Feb 1997 19:12:59 -0800 (PST)
Message-Id: <199702250312.TAA02667@silvia.HIP.Berkeley.EDU>
To: mark@quickweb.com
CC: kuku@gilberto.physik.rwth-aachen.de, robsch@robkaos.ruhr.de,
        freebsd-hardware@freebsd.org
In-reply-to: <Pine.BSF.3.95q.970224210037.6213D-100000@vinyl.quickweb.com> (message from Mark Mayo on Mon, 24 Feb 1997 21:02:35 -0500 (EST))
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 * From: Mark Mayo <mark@quickweb.com>

 * > > 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
 * > > The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.

 * > This is from my P6NP5:
 * > 2097152000 bytes transferred in 25.741665 secs (81469167 bytes/sec)      
 * 
 * While we're at it, here are the results fom my Intel 440FX(Natoma) P6-200
 * 2097152000 bytes transferred in 24 secs (87381333 bytes/sec)
 * 
 * My first generation PPro 150 from Digital only musters ~60MB/s..

Hmm, how come my new P6NP5 with P6-200 and 96MB RAM (parity mode) can
give me only about:

>> dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 18.359226 secs (57114390 bytes/sec)

I tried fiddling with the "RAS precharge" and "RAS to CAS delay" in
the BIOS setups but the numbers hardly changed (always between 57MB/s
and 60MB/s).

Did I get a "first generation" 440FX or something? ;(

Satoshi

From owner-freebsd-hardware  Mon Feb 24 20:20:34 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id UAA19795
          for hardware-outgoing; Mon, 24 Feb 1997 20:20:34 -0800 (PST)
Received: from george.lbl.gov (george.lbl.gov [128.3.196.93])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id UAA19780
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 20:20:26 -0800 (PST)
Received: from portnoy.lbl.gov (portnoy.lbl.gov [131.243.2.11]) by george.lbl.gov (8.6.10/8.6.5) with ESMTP id UAA15881; Mon, 24 Feb 1997 20:15:27 -0800
From: "Jin Guojun[ITG]" <jin@george.lbl.gov>
Received: (jin@localhost) by portnoy.lbl.gov (8.6.10/8.6.5) id UAA09562; Mon, 24 Feb 1997 20:20:12 -0800
Date: Mon, 24 Feb 1997 20:20:12 -0800
Message-Id: <199702250420.UAA09562@portnoy.lbl.gov>
To: asami@vader.cs.berkeley.edu, mark@quickweb.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>  * From: Mark Mayo <mark@quickweb.com>
> 
>  * > > 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
>  * > > The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
> 
>  * > This is from my P6NP5:
>  * > 2097152000 bytes transferred in 25.741665 secs (81469167 bytes/sec)      
>  * 
>  * While we're at it, here are the results fom my Intel 440FX(Natoma) P6-200
>  * 2097152000 bytes transferred in 24 secs (87381333 bytes/sec)
>  * 
>  * My first generation PPro 150 from Digital only musters ~60MB/s..
> 
> Hmm, how come my new P6NP5 with P6-200 and 96MB RAM (parity mode) can
> give me only about:
> 
> >> dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 18.359226 secs (57114390 bytes/sec)
> 
> I tried fiddling with the "RAS precharge" and "RAS to CAS delay" in
> the BIOS setups but the numbers hardly changed (always between 57MB/s
> and 60MB/s).
> 
> Did I get a "first generation" 440FX or something? ;(
> 
> Satoshi

Do not waste time to play this game. The "dd" is O.S. dependent code.
It does not give you what is real memory speed on your system. The result
from dd is really depended on the O.S. you are running. If you run 2.2 or
higher, you will get much better performance than 2.1.x.

440FX does have worse memory speed than Triton-{I,II}; even though P6 has
much better CPU speed, but the PCI controller (440FX) is worse.

hopefully this will more be clear.

-Jin


From owner-freebsd-hardware  Mon Feb 24 21:04:22 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id VAA23138
          for hardware-outgoing; Mon, 24 Feb 1997 21:04:22 -0800 (PST)
Received: from minor.stranger.com (stranger.vip.best.com [204.156.129.250])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id VAA23107
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 21:04:02 -0800 (PST)
Received: from dog.farm.org (dog.farm.org [207.111.140.47]) by minor.stranger.com (8.6.12/8.6.12) with ESMTP id VAA28775 for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 21:24:42 -0800
Received: (from dk@localhost) by dog.farm.org (8.7.5/dk#3) id VAA15215 for freebsd-hardware@freebsd.org; Mon, 24 Feb 1997 21:07:25 -0800 (PST)
Date: Mon, 24 Feb 1997 21:07:25 -0800 (PST)
From: Dmitry Kohmanyuk <dk@dog.farm.org>
Message-Id: <199702250507.VAA15215@dog.farm.org>
Subject: Re: multi-interface ethernet cards
Cc: freebsd-hardware@freebsd.org
Newsgroups: cs-monolit.gated.lists.freebsd.hardware
References: <199702242335.NAA22803@pegasus.com>
Organization: FARM Computing Association
Reply-To: dk+@ua.net
X-Newsreader: TIN [version 1.2 PL2]
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Richard Foulk (richard@pegasus.com) wrote:
> } > Does any know if there are ethernet cards available (and supported) with
> } > multiple interfaces (like 4).
> } 
> } The Znyx cards have 4 ethernets and as they use the DEC chipset, we support
> } them.
> } -- 

> What are the prices like on these cards and where can you order them from?

I strongly recommend ASA Computers (www.asacomputers.com).  They are in SJ, CA.

Znyx 314 (4-port 10baseT, PCI) is avail from them immediately;  prices
are on the Web site.  Around 500, if I remember correctly...

Their phone is on their Web site.  Ask Kedar - he can make a FreeBSD-compatible
machine, pre-install FreeBSD for you, build a custom kernel and setup X11.  

--
 If you lie to the compiler, it will get its revenge.
					   - Henry Spencer

From owner-freebsd-hardware  Mon Feb 24 22:01:49 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id WAA26472
          for hardware-outgoing; Mon, 24 Feb 1997 22:01:49 -0800 (PST)
Received: from who.cdrom.com (who.cdrom.com [204.216.27.3])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA26449
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 22:01:42 -0800 (PST)
Received: from stox.pr.mcs.net (stox.pr.mcs.net [204.137.243.33])
          by who.cdrom.com (8.7.5/8.6.11) with ESMTP id VAA10786
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 21:45:59 -0800 (PST)
Received: from localhost (localhost.stox.pr.mcs.net [127.0.0.1]) by stox.pr.mcs.net (8.8.5/8.6.12) with SMTP id XAA23728 for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 23:44:50 -0600 (CST)
Date: Mon, 24 Feb 1997 23:44:49 -0600 (CST)
From: "Kenneth P. Stox" <ken@stox.pr.mcs.net>
To: freebsd-hardware@freebsd.org
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <199702250312.TAA02667@silvia.HIP.Berkeley.EDU>
Message-ID: <Pine.BSI.3.95.970224232426.9325E-100000@stox.pr.mcs.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


On Mon, 24 Feb 1997, Satoshi Asami wrote:

>  * From: Mark Mayo <mark@quickweb.com>
> 
>  * > > 2097152000 bytes transferred in 27.298230 secs (76823735 bytes/sec)
>  * > > The motherboard is a P6NP5 (Natoma chipset) with 64 MB EDO-RAM.
> 
>  * > This is from my P6NP5:
>  * > 2097152000 bytes transferred in 25.741665 secs (81469167 bytes/sec)      
>  * 
>  * While we're at it, here are the results fom my Intel 440FX(Natoma) P6-200
>  * 2097152000 bytes transferred in 24 secs (87381333 bytes/sec)
>  * 
>  * My first generation PPro 150 from Digital only musters ~60MB/s..
> 
> Hmm, how come my new P6NP5 with P6-200 and 96MB RAM (parity mode) can
> give me only about:
> 
> >> dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 18.359226 secs (57114390 bytes/sec)

Double Hmmm....... My P5STE (512KB Cache, 430HX)  with a P5-120 and 64MB
EDO RAM, 3.0 current as of 2/10/97:

dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 10.875472 secs (96416596 bytes/sec)

Am I seeing things ?

-Ken Stox
 ken@stox.pr.mcs.net


From owner-freebsd-hardware  Mon Feb 24 22:58:04 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id WAA01910
          for hardware-outgoing; Mon, 24 Feb 1997 22:58:04 -0800 (PST)
Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA01905
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 22:57:58 -0800 (PST)
Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id RAA13458; Tue, 25 Feb 1997 17:26:35 +1030 (CST)
From: Michael Smith <msmith@atrad.adelaide.edu.au>
Message-Id: <199702250656.RAA13458@genesis.atrad.adelaide.edu.au>
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <Pine.BSI.3.95.970224232426.9325E-100000@stox.pr.mcs.net> from "Kenneth P. Stox" at "Feb 24, 97 11:44:49 pm"
To: ken@stox.pr.mcs.net (Kenneth P. Stox)
Date: Tue, 25 Feb 1997 17:26:29 +1030 (CST)
Cc: freebsd-hardware@freebsd.org
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Kenneth P. Stox stands accused of saying:
> 
> Double Hmmm....... My P5STE (512KB Cache, 430HX)  with a P5-120 and 64MB
> EDO RAM, 3.0 current as of 2/10/97:
> 
> dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 10.875472 secs (96416596 bytes/sec)
> 
> Am I seeing things ?

No, that's about right.  You will have the i586-optimised
copyin/copyout code, which makes the P5 faster (on this particular
benchmark) than the P6.  This is a P5/166 with 60ns ECC (_not_ EDO) DRAM,
512K cache, 430HX and conservative BIOS settings :

bom:~>dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 8.186862 secs (128080332 bytes/sec)

As Bruce has observed, this makes for very healthy IDE performance,
which is what this whole thread started with.

> -Ken Stox

-- 
]] Mike Smith, Software Engineer        msmith@gsoft.com.au             [[
]] Genesis Software                     genesis@gsoft.com.au            [[
]] High-speed data acquisition and      (GSM mobile)     0411-222-496   [[
]] realtime instrument control.         (ph)          +61-8-8267-3493   [[
]] Unix hardware collector.             "Where are your PEZ?" The Tick  [[

From owner-freebsd-hardware  Mon Feb 24 23:30:10 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id XAA04675
          for hardware-outgoing; Mon, 24 Feb 1997 23:30:10 -0800 (PST)
Received: from dfw-ix7.ix.netcom.com (dfw-ix7.ix.netcom.com [206.214.98.7])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA04670
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 23:30:07 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix7.ix.netcom.com (8.8.4/8.8.4)
	  id BAA22273; Tue, 25 Feb 1997 01:29:01 -0600 (CST)
Received: from ala-ca13-23.ix.netcom.com(204.32.168.55) by dfw-ix7.ix.netcom.com via smap (V1.3)
	id sma021932; Tue Feb 25 01:28:49 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id XAA05308; Mon, 24 Feb 1997 23:28:08 -0800 (PST)
Date: Mon, 24 Feb 1997 23:28:08 -0800 (PST)
Message-Id: <199702250728.XAA05308@silvia.HIP.Berkeley.EDU>
To: jin@george.lbl.gov
CC: mark@quickweb.com, freebsd-hardware@freebsd.org,
        kuku@gilberto.physik.rwth-aachen.de, robsch@robkaos.ruhr.de
In-reply-to: <199702250420.UAA09562@portnoy.lbl.gov> (jin@george.lbl.gov)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 * > Hmm, how come my new P6NP5 with P6-200 and 96MB RAM (parity mode) can
 * > give me only about:
 * > 
 * > >> dd if=/dev/zero of=/dev/null bs=1m count=1000
 * > 1000+0 records in
 * > 1000+0 records out
 * > 1048576000 bytes transferred in 18.359226 secs (57114390 bytes/sec)

 * Do not waste time to play this game. The "dd" is O.S. dependent code.

Sorry I didn't write the OS version; I am running 3.0 from just before 
the Lite/2 merge.

 * It does not give you what is real memory speed on your system. The result
 * from dd is really depended on the O.S. you are running. If you run 2.2 or
 * higher, you will get much better performance than 2.1.x.

Well I wouldn't ask the list if it's just dd but I've run several
memory tests of my own and this machine always comes out looking very
bad compared to the P5-133 (Triton-II) it replaced.  It's always about 
half, unless the whole thing fits in the 512K L2 cache (in which case
it smokes).

 * 440FX does have worse memory speed than Triton-{I,II}; even though P6 has
 * much better CPU speed, but the PCI controller (440FX) is worse.

I know that.  But there are some people seeing 80MB/s or more and some 
(including myself) who only get about 60MB/s on an apparently
identical chipset.

Satoshi

From owner-freebsd-hardware  Mon Feb 24 23:33:05 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id XAA04905
          for hardware-outgoing; Mon, 24 Feb 1997 23:33:05 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA04900
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 23:33:00 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id SAA00656; Tue, 25 Feb 1997 18:30:56 +1100
Date: Tue, 25 Feb 1997 18:30:56 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702250730.SAA00656@godzilla.zeta.org.au>
To: asami@vader.cs.berkeley.edu, jin@george.lbl.gov, mark@quickweb.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>Do not waste time to play this game. The "dd" is O.S. dependent code.

No, dd is very machine-independent.  It just loops calling read() and
write() with the specified block size.  However, the implementation
of /dev/zero is very machine-dependent.  FreeBSD happens to have an
implementation that copies memory in a straightforward way, so the speed
reported by dd is closely related to the memory write bandwidth.  The
read bandwidth doesn't matter much because most reads are from the
cache.

>It does not give you what is real memory speed on your system. The result
>from dd is really depended on the O.S. you are running. If you run 2.2 or
>higher, you will get much better performance than 2.1.x.

There isn't much difference unless you have a P5 and the P5-optimized
copyout routine is not disabled.

>440FX does have worse memory speed than Triton-{I,II}; even though P6 has
>much better CPU speed, but the PCI controller (440FX) is worse.

That would explain why a P6 is more than twice as slow as a P5 for large
writes.  It is inherently about twice as slow because P6's read cache
lines before writing to them.  Read-before-write is useful if the memory
is read after writing to it, but the dd benchmark never does that.

Bruce

From owner-freebsd-hardware  Mon Feb 24 23:57:12 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id XAA07216
          for hardware-outgoing; Mon, 24 Feb 1997 23:57:12 -0800 (PST)
Received: from dfw-ix9.ix.netcom.com (dfw-ix9.ix.netcom.com [206.214.98.9])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA07211
          for <freebsd-hardware@freebsd.org>; Mon, 24 Feb 1997 23:57:09 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix9.ix.netcom.com (8.8.4/8.8.4)
	  id BAA27415 for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 01:56:37 -0600 (CST)
Received: from ala-ca13-23.ix.netcom.com(204.32.168.55) by dfw-ix9.ix.netcom.com via smap (V1.3)
	id sma024577; Tue Feb 25 01:34:52 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id XAA05342; Mon, 24 Feb 1997 23:33:00 -0800 (PST)
Date: Mon, 24 Feb 1997 23:33:00 -0800 (PST)
Message-Id: <199702250733.XAA05342@silvia.HIP.Berkeley.EDU>
To: ken@stox.pr.mcs.net
CC: freebsd-hardware@freebsd.org
In-reply-to: <Pine.BSI.3.95.970224232426.9325E-100000@stox.pr.mcs.net> (ken@stox.pr.mcs.net)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 * From: "Kenneth P. Stox" <ken@stox.pr.mcs.net>

 * Double Hmmm....... My P5STE (512KB Cache, 430HX)  with a P5-120 and 64MB
 * EDO RAM, 3.0 current as of 2/10/97:
 * 
 * dd if=/dev/zero of=/dev/null bs=1m count=1000
 * 1000+0 records in
 * 1000+0 records out
 * 1048576000 bytes transferred in 10.875472 secs (96416596 bytes/sec)
 * 
 * Am I seeing things ?

That sounds just about right for a P5-120 (note the 60MHz external
clock) when compared with 115-120MB/s for P5-133 that other people
(including myself) have been seeing.

Satoshi

From owner-freebsd-hardware  Tue Feb 25 00:11:36 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA08213
          for hardware-outgoing; Tue, 25 Feb 1997 00:11:36 -0800 (PST)
Received: from kremvax.demos.su (kremvax.demos.su [194.87.0.20])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA08204
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 00:11:30 -0800 (PST)
Received: by kremvax.demos.su (8.6.13/D) from 0@sinbin.demos.su [194.87.2.95]
          with ESMTP id LAA07064; Tue, 25 Feb 1997 11:05:57 +0300
Received: by sinbin.demos.su id LAA17129;
  (8.6.12/D) Tue, 25 Feb 1997 11:05:22 +0300
From: bag@sinbin.demos.su (Alex G. Bulushev)
Message-Id: <199702250805.LAA17129@sinbin.demos.su>
Subject: Re: Memory speed of P6-200 (256k)
To: asami@vader.cs.berkeley.edu (Satoshi Asami)
Date: Tue, 25 Feb 1997 11:05:22 +0300 (MSK)
Cc: jin@george.lbl.gov, mark@quickweb.com, freebsd-hardware@freebsd.org,
        kuku@gilberto.physik.rwth-aachen.de, robsch@robkaos.ruhr.de
In-Reply-To: <199702250728.XAA05308@silvia.HIP.Berkeley.EDU> from "Satoshi Asami" at Feb 24, 97 11:28:08 pm
X-Mailer: ELM [version 2.4 PL24 ME7a]
Content-Type: text
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
>  * > Hmm, how come my new P6NP5 with P6-200 and 96MB RAM (parity mode) can
>  * > give me only about:
>  * > 
>  * > >> dd if=/dev/zero of=/dev/null bs=1m count=1000
>  * > 1000+0 records in
>  * > 1000+0 records out
>  * > 1048576000 bytes transferred in 18.359226 secs (57114390 bytes/sec)

Asus P/I-P65UP5 with C-P6ND  single P6-200 256k and 256MB RAM EDO
fbsd 2.2-GAMMA

dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 12.738651 secs (82314524 bytes/sec)

     Alex.


> 
>  * Do not waste time to play this game. The "dd" is O.S. dependent code.
> 
> Sorry I didn't write the OS version; I am running 3.0 from just before 
> the Lite/2 merge.
> 
>  * It does not give you what is real memory speed on your system. The result
>  * from dd is really depended on the O.S. you are running. If you run 2.2 or
>  * higher, you will get much better performance than 2.1.x.
> 
> Well I wouldn't ask the list if it's just dd but I've run several
> memory tests of my own and this machine always comes out looking very
> bad compared to the P5-133 (Triton-II) it replaced.  It's always about 
> half, unless the whole thing fits in the 512K L2 cache (in which case
> it smokes).
> 
>  * 440FX does have worse memory speed than Triton-{I,II}; even though P6 has
>  * much better CPU speed, but the PCI controller (440FX) is worse.
> 
> I know that.  But there are some people seeing 80MB/s or more and some 
> (including myself) who only get about 60MB/s on an apparently
> identical chipset.
> 
> Satoshi
> 


From owner-freebsd-hardware  Tue Feb 25 00:22:57 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA08680
          for hardware-outgoing; Tue, 25 Feb 1997 00:22:57 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA08675
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 00:22:54 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id TAA02011; Tue, 25 Feb 1997 19:19:41 +1100
Date: Tue, 25 Feb 1997 19:19:41 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702250819.TAA02011@godzilla.zeta.org.au>
To: ken@stox.pr.mcs.net, msmith@atrad.adelaide.edu.au
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>As Bruce has observed, this makes for very healthy IDE performance,
>which is what this whole thread started with.

Er, no, fast copying reduces buffer cache overheads.  Buffer cache
overheads are almost large enough to significantly reduce the advantages
of busmastering DMA over (slow) PIO.

Bruce

From owner-freebsd-hardware  Tue Feb 25 00:29:18 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA09642
          for hardware-outgoing; Tue, 25 Feb 1997 00:29:18 -0800 (PST)
Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA09630
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 00:29:14 -0800 (PST)
Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id SAA14217; Tue, 25 Feb 1997 18:59:01 +1030 (CST)
From: Michael Smith <msmith@atrad.adelaide.edu.au>
Message-Id: <199702250829.SAA14217@genesis.atrad.adelaide.edu.au>
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <199702250819.TAA02011@godzilla.zeta.org.au> from Bruce Evans at "Feb 25, 97 07:19:41 pm"
To: bde@zeta.org.au (Bruce Evans)
Date: Tue, 25 Feb 1997 18:58:59 +1030 (CST)
Cc: ken@stox.pr.mcs.net, msmith@atrad.adelaide.edu.au,
        freebsd-hardware@freebsd.org
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Bruce Evans stands accused of saying:
> >As Bruce has observed, this makes for very healthy IDE performance,
> >which is what this whole thread started with.
> 
> Er, no, fast copying reduces buffer cache overheads.  Buffer cache
> overheads are almost large enough to significantly reduce the advantages
> of busmastering DMA over (slow) PIO.

ie. the fast copyin/out code will make busmaster devices faster too? 8)

> Bruce

-- 
]] Mike Smith, Software Engineer        msmith@gsoft.com.au             [[
]] Genesis Software                     genesis@gsoft.com.au            [[
]] High-speed data acquisition and      (GSM mobile)     0411-222-496   [[
]] realtime instrument control.         (ph)          +61-8-8267-3493   [[
]] Unix hardware collector.             "Where are your PEZ?" The Tick  [[

From owner-freebsd-hardware  Tue Feb 25 01:21:59 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id BAA12512
          for hardware-outgoing; Tue, 25 Feb 1997 01:21:59 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA12507
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 01:21:56 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id UAA03535; Tue, 25 Feb 1997 20:18:42 +1100
Date: Tue, 25 Feb 1997 20:18:42 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702250918.UAA03535@godzilla.zeta.org.au>
To: bde@zeta.org.au, msmith@atrad.adelaide.edu.au
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, ken@stox.pr.mcs.net
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>> Er, no, fast copying reduces buffer cache overheads.  Buffer cache
>> overheads are almost large enough to significantly reduce the advantages
>> of busmastering DMA over (slow) PIO.
>
>ie. the fast copyin/out code will make busmaster devices faster too? 8)

Yes, that was the original motivation for speeding up copyout() and
copyin().  Satoshi had lots of disk bandwidth (40MB/sec?) from multiple
controllers but couldn't use it all because copying alone was limited
to 40MB/sec.  He speeded it up to 70+MB/sec on a P5-Triton system by
using the FPU, and I speeded it up a few more MB/sec by fine tuning.
The dd speeds are almost twice as large since the input buffer is small
enough to stay in the P5's L1 data cache.

Bruce

From owner-freebsd-hardware  Tue Feb 25 11:53:05 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id LAA11621
          for hardware-outgoing; Tue, 25 Feb 1997 11:53:05 -0800 (PST)
Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA11610
          for <hardware@freebsd.org>; Tue, 25 Feb 1997 11:52:59 -0800 (PST)
Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id MAA23943; Tue, 25 Feb 1997 12:52:52 -0700 (MST)
Date: Tue, 25 Feb 1997 12:52:52 -0700 (MST)
Message-Id: <199702251952.MAA23943@rocky.mt.sri.com>
From: Nate Williams <nate@mt.sri.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: Nate Williams <nate@mt.sri.com>
Cc: Bruce Evans <bde@zeta.org.au>, hardware@freebsd.org
Subject: Re: Memory speed (was Re: _big_ IDE disks?)
In-Reply-To: <199702231629.JAA11518@rocky.mt.sri.com>
References: <199702231441.BAA01194@godzilla.zeta.org.au>
	<199702231629.JAA11518@rocky.mt.sri.com>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> FreeBSD 2.2-GAMMA #10: Fri Feb 21 18:07:37 MST 1997
> CPU: Pentium (99.47-MHz 586-class CPU)
>   Origin = "GenuineIntel"  Id = 0x525  Stepping=5
>   Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
> real memory  = 67108864 (65536K bytes)
> avail memory = 63631360 (62140K bytes)
> 
>  *      $Id: npx.c,v 1.31.2.6 1997/02/13 06:59:51 bde Exp $
> 
> moth:/sys/i386/isa # dd if=/dev/zero of=/dev/null bs=1m count=1000
> 1000+0 records in
> 1000+0 records out
> 1048576000 bytes transferred in 14.549281 secs (72070640 bytes/sec)
> 
> Pretty low.  This is an ASUS/Triton-I board if it makes any difference.

I just upgraded it to a 166.

FreeBSD 2.2-GAMMA #10: Fri Feb 21 18:07:37 MST 1997
    root@moth.mt.sri.com:/usr/src/sys/compile/MOTH
Calibrating clock(s) relative to mc146818A clock ... i586 clock: 165778631 Hz, i
8254 clock: 1193106 Hz
CPU: Pentium (165.79-MHz 586-class CPU)
  Origin = "GenuineIntel"  Id = 0x52c  Stepping=12
  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 67108864 (65536K bytes)
avail memory = 63631360 (62140K bytes)

moth:/sys/compile/MOTH # dd if=/dev/zero of=/dev/null count=1000 bs=1m
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 8.264108 secs (126883142 bytes/sec)

Big difference, so it's gotta be related to the CPU speed and not just
the CPU class.



Nate

From owner-freebsd-hardware  Tue Feb 25 12:53:58 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id MAA15375
          for hardware-outgoing; Tue, 25 Feb 1997 12:53:58 -0800 (PST)
Received: from george.lbl.gov (george.lbl.gov [128.3.196.93])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA15364
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 12:53:53 -0800 (PST)
Received: (jin@localhost) by george.lbl.gov (8.6.10/8.6.5) id MAA06713; Tue, 25 Feb 1997 12:48:51 -0800
Date: Tue, 25 Feb 1997 12:48:51 -0800
From: "Jin Guojun[ITG]" <jin@george.lbl.gov>
Message-Id: <199702252048.MAA06713@george.lbl.gov>
To: asami@vader.cs.berkeley.edu, bde@zeta.org.au, mark@quickweb.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

} >Do not waste time to play this game. The "dd" is O.S. dependent code.
} 
} No, dd is very machine-independent.  It just loops calling read() and
} write() with the specified block size.  However, the implementation
} of /dev/zero is very machine-dependent.  FreeBSD happens to have an
} implementation that copies memory in a straightforward way, so the speed
} reported by dd is closely related to the memory write bandwidth.  The
} read bandwidth doesn't matter much because most reads are from the
} cache.
} 
} >It does not give you what is real memory speed on your system. The result
} >from dd is really depended on the O.S. you are running. If you run 2.2 or
} >higher, you will get much better performance than 2.1.x.
} 
} There isn't much difference unless you have a P5 and the P5-optimized
} copyout routine is not disabled.
} 
} Bruce

% dmesg
FreeBSD 2.1.7-RELEASE #0: Thu Feb 20 20:44:03 PST 1997
    root@adv-pc-1.lbl.gov:/usr/src/sys/compile/MinMax
CPU: 200-MHz Pentium 735\\90 or 815\\100 (Pentium-class CPU)
  Origin = "GenuineIntel"  Id = 0x52c  Stepping=12
  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 67108864 (65536K bytes)
avail memory = 63025152 (61548K bytes)
...
% dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 12 secs (87381333 bytes/sec)
0.0u 12.3s 0:12.43 99.4% 51+2809k 0+0io 3pf+0w
### this result matches the standard memory bandwidth

*** The same machine with different FreeBSD ***

% dmesg 
Copyright (c) 1992-1996 FreeBSD Inc.
Copyright (c) 1982, 1986, 1989, 1991, 1993
        The Regents of the University of California.  All rights reserved.

FreeBSD 2.2-970215-GAMMA #0: Wed Feb 19 15:22:41 PST 1997
    root@adv-pc-1.lbl.gov:/usr/src/sys/compile/MinMax
Calibrating clock(s) relative to mc146818A clock ... i586 clock: 200455533 Hz, i8254 clock: 1193190 Hz
CPU: Pentium (200.45-MHz 586-class CPU)
  Origin = "GenuineIntel"  Id = 0x52c  Stepping=12
  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
real memory  = 67108864 (65536K bytes)
avail memory = 62623744 (61156K bytes)

% dd if=/dev/zero of=/dev/null bs=1m count=1000
1000+0 records in
1000+0 records out
1048576000 bytes transferred in 7.591523 secs (138124591 bytes/sec)
0.0u 7.5s 0:07.62 99.4% 71+2841k 0+0io 3pf+0w

# this result is better than the standard memory bandwidth, but worse than
the maximum memory bandwidth.

That is why I said "It is O.S. dependent."

} ----------------------------------------------------------------------
}  * 440FX does have worse memory speed than Triton-{I,II}; even though P6 has
}  * much better CPU speed, but the PCI controller (440FX) is worse.
} 
} I know that.  But there are some people seeing 80MB/s or more and some 
} (including myself) who only get about 60MB/s on an apparently
} identical chipset.
} 
} Satoshi

So, as we discussed before, you should notice that the result from "dd" does
not show what memory speed you really can get from your system, unless you
specifically use dd only.

The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).
However, no memory sub-system in PC/UNIX O.S. uses 64-bit memory bandwidth
currently because of the CPU bus.
As I saw in the other message, you use FPU to achive this goal,
which is what you can see in ftp://george.lbl.gov/pub/ccs/performance.ps,
the result of 8-byte register to memory copy. This is what you really can get.

Another tip is even some motherboards are using same PCI shipset, the memory
performance may vary.
For example, compare ASUS Triton-{I, II} with three different Intel
motherboards (EV2, PT-2000, ZAPPA) with Triton-II PCI chipset,
the memory I/O speed does not have much different;
however, ASUS motherboard can use 70ns memory chip v.s. Intel motherboards have
to use 60ns memory chip. Very unbelieveable fact, right? This is hardware issue.

The software issue is that all memory sub-system (including string system) 
are written in assemble language (NOT in C). The qulitiy of this piece of code
is critical to the memory performance. So, doing cross O.S. memory performance
comparssion is meaningless.

By understanding the memory system to determine the its speed is helpful.
So, to get 67 MBps from current PCI bus is normal. To get more than this
speed is feasible, depends on how to play the trick in the memory system.

-Jin


From owner-freebsd-hardware  Tue Feb 25 15:16:05 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id PAA23217
          for hardware-outgoing; Tue, 25 Feb 1997 15:16:05 -0800 (PST)
Received: from mail12.digital.com (mail12.digital.com [192.208.46.20])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA23191
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 15:16:00 -0800 (PST)
Received: from muggsy.lkg.dec.com by mail12.digital.com (8.7.5/UNX 1.5/1.0/WV)
	id SAA15169; Tue, 25 Feb 1997 18:06:10 -0500 (EST)
Received:  from usr602.zko.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP
	id AA25703; Tue, 25 Feb 1997 18:06:07 -0500
Message-Id: <3.0.32.19691231190000.0069517c@netrix.lkg.dec.com>
X-Sender: popmatt@netrix.lkg.dec.com
X-Mailer: Windows Eudora Pro Version 3.0 Demo (32)
Date: Tue, 25 Feb 1997 18:06:18 -0500
To: "Jin Guojun[ITG]" <jin@george.lbl.gov>
From: Matt Thomas <matt@lkg.dec.com>
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

At 12:48 PM 2/25/97 -0800, Jin Guojun[ITG] wrote:
>The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
>this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).

That is wrong.  The PCI is typically a 32-bit wide bus running @ up to 33Mhz.

4 bytes * 33Mhz = maximum of 132MB/s assuming almost perfect bursting
and no wait states.

64 bit PCI doubles that maximum to 265MB/s.
-- 
Matt Thomas               Internet:   matt@3am-software.com
3am Software Foundry      WWW URL:    http://www.3am-software.com/bio/matt.html
Westford, MA              Disclaimer: I disavow all knowledge of this message


From owner-freebsd-hardware  Tue Feb 25 15:58:02 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id PAA25678
          for hardware-outgoing; Tue, 25 Feb 1997 15:58:02 -0800 (PST)
Received: from george.lbl.gov (george.lbl.gov [128.3.196.93])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA25673
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 15:57:59 -0800 (PST)
Received: (jin@localhost) by george.lbl.gov (8.6.10/8.6.5) id PAA12462; Tue, 25 Feb 1997 15:53:06 -0800
Date: Tue, 25 Feb 1997 15:53:06 -0800
From: "Jin Guojun[ITG]" <jin@george.lbl.gov>
Message-Id: <199702252353.PAA12462@george.lbl.gov>
To: matt@lkg.dec.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

} At 12:48 PM 2/25/97 -0800, Jin Guojun[ITG] wrote:
} >The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
} >this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).
} 
} That is wrong.  The PCI is typically a 32-bit wide bus running @ up to 33Mhz.
} 
} 4 bytes * 33Mhz = maximum of 132MB/s assuming almost perfect bursting
} and no wait states.
} 
} 64 bit PCI doubles that maximum to 265MB/s.

That is the PCI bus bandwidth, not the memory bandwidth :-)
PCI != MEOROY, but main memory uses PCI bus.

-Jin


From owner-freebsd-hardware  Tue Feb 25 17:59:59 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id RAA04427
          for hardware-outgoing; Tue, 25 Feb 1997 17:59:59 -0800 (PST)
Received: from mail11.digital.com (mail11.digital.com [192.208.46.10])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA04414
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 17:59:53 -0800 (PST)
Received: from muggsy.lkg.dec.com by mail11.digital.com (8.7.5/UNX 1.5/1.0/WV)
	id UAA30486; Tue, 25 Feb 1997 20:52:25 -0500 (EST)
Received:  from usr605.zko.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP
	id AA26709; Tue, 25 Feb 1997 20:52:22 -0500
Message-Id: <3.0.32.19970225205045.006884cc@netrix.lkg.dec.com>
X-Sender: popmatt@netrix.lkg.dec.com
X-Mailer: Windows Eudora Pro Version 3.0 Demo (32)
Date: Tue, 25 Feb 1997 20:52:37 -0500
To: "Jin Guojun[ITG]" <jin@george.lbl.gov>
From: Matt Thomas <matt@lkg.dec.com>
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

At 03:53 PM 2/25/97 -0800, Jin Guojun[ITG] wrote:
>} At 12:48 PM 2/25/97 -0800, Jin Guojun[ITG] wrote:
>} >The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
>} >this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).
>} 
>} That is wrong.  The PCI is typically a 32-bit wide bus running @ up to 33Mhz.
>} 
>} 4 bytes * 33Mhz = maximum of 132MB/s assuming almost perfect bursting
>} and no wait states.
>} 
>} 64 bit PCI doubles that maximum to 265MB/s.
>
>That is the PCI bus bandwidth, not the memory bandwidth :-)
>PCI != MEOROY, but main memory uses PCI bus.

The main memory access may be done through a PCI host bridge, but that
does not mean that's it done through the PCI.  A system/cpu may have
more than one PCI bus (like Alpha's or high-end Pentium Pro's).  
Memory bandwidth is not directly related to PCI bandwidth.
-- 
Matt Thomas               Internet:   matt@3am-software.com
3am Software Foundry      WWW URL:    http://www.3am-software.com/bio/matt.html
Westford, MA              Disclaimer: I disavow all knowledge of this message


From owner-freebsd-hardware  Tue Feb 25 21:46:14 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id VAA16778
          for hardware-outgoing; Tue, 25 Feb 1997 21:46:14 -0800 (PST)
Received: from dfw-ix15.ix.netcom.com (dfw-ix15.ix.netcom.com [206.214.98.15])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA16773
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 21:46:11 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix15.ix.netcom.com (8.8.4/8.8.4)
	  id XAA09414; Tue, 25 Feb 1997 23:44:47 -0600 (CST)
Received: from wck-ca5-07.ix.netcom.com(199.35.213.167) by dfw-ix15.ix.netcom.com via smap (V1.3)
	id sma009357; Tue Feb 25 23:44:27 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id VAA01458; Tue, 25 Feb 1997 21:44:19 -0800 (PST)
Date: Tue, 25 Feb 1997 21:44:19 -0800 (PST)
Message-Id: <199702260544.VAA01458@silvia.HIP.Berkeley.EDU>
To: bde@zeta.org.au
CC: bde@zeta.org.au, msmith@atrad.adelaide.edu.au,
        freebsd-hardware@freebsd.org, ken@stox.pr.mcs.net
In-reply-to: <199702250918.UAA03535@godzilla.zeta.org.au> (message from Bruce Evans on Tue, 25 Feb 1997 20:18:42 +1100)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

I finally found what was wrong.  It was pilot error (shame on me! ;).

When I bought the motherboard, I went to the BIOS setup and called the 
menu item to reset everything to the default (like any smart boy will
do).  Well, the only problem was that I wasn't paying much attention
and didn't notice that there were two "default"s -- the "BIOS
defaults" and "setup defaults".  According to the manual (that I
finally read from beginning to end today), the "real" (optimized)
defaults are the latter; the first one is a fail-safe, debugging-mode
type of thing.  Duh.

So, I invoked the "setup defaults" and voila! I got >80MB/s.  I even
played around with the memory timings and managed to add a few MB/s.
(As Chris said, changing the "RAS Precharge" from 3T to 4T helped.)
Now I see something like:

>> dd if=/dev/zero of=/dev/null bs=1m count=100
100+0 records in
100+0 records out
104857600 bytes transferred in 1.198974 secs (87456107 bytes/sec)

There is no difference at all whether it's in ECC mode, parity mode or 
they are both disabled.  Quite interesting.

Anyway, so I'm happy now.  People, thanks for all the encouragement. :)

 * Yes, that was the original motivation for speeding up copyout() and
 * copyin().  Satoshi had lots of disk bandwidth (40MB/sec?) from multiple

More like >60MB/s, we had four F-W SCSI strings at one point.

 * controllers but couldn't use it all because copying alone was limited
 * to 40MB/sec.  He speeded it up to 70+MB/sec on a P5-Triton system by
 * using the FPU, and I speeded it up a few more MB/sec by fine tuning.

Yes.  With the new 80MB/s copyin/out, the max throughput of reads
of a ccd through the filesystem jumped from 21MB/s to 27MB/s using 4
Seagate Barracuda 15150WC's.

Satoshi

From owner-freebsd-hardware  Tue Feb 25 22:34:07 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id WAA19804
          for hardware-outgoing; Tue, 25 Feb 1997 22:34:07 -0800 (PST)
Received: from dfw-ix12.ix.netcom.com (dfw-ix12.ix.netcom.com [206.214.98.12])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA19797
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 22:34:02 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix12.ix.netcom.com (8.8.4/8.8.4)
	  id AAA17253; Wed, 26 Feb 1997 00:32:57 -0600 (CST)
Received: from wck-ca5-07.ix.netcom.com(199.35.213.167) by dfw-ix12.ix.netcom.com via smap (V1.3)
	id sma017097; Wed Feb 26 00:31:46 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id WAA01627; Tue, 25 Feb 1997 22:31:36 -0800 (PST)
Date: Tue, 25 Feb 1997 22:31:36 -0800 (PST)
Message-Id: <199702260631.WAA01627@silvia.HIP.Berkeley.EDU>
To: matt@lkg.dec.com
CC: jin@george.lbl.gov, freebsd-hardware@freebsd.org
In-reply-to: <3.0.32.19970225205045.006884cc@netrix.lkg.dec.com> (message from Matt Thomas on Tue, 25 Feb 1997 20:52:37 -0500)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

 * >} At 12:48 PM 2/25/97 -0800, Jin Guojun[ITG] wrote:
 * >} >The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
 * >} >this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).

This is totally bogus.

The "60ns" is just the row access time, i.e., the time it takes for
the data lines to stabilize after you send a request to the memory.
There is much more in memory bandwidth than just row access time.

For starters, there is fast page mode (FPM: get more words from the
same row in memory for only about half the time for row access) and
cycle time (the interval between accesses to different rows usually
require about twice the row access time).

So, for instance, if your memory system and chipset transfers 64 bits
at a time (like P5 or P6), and you can get 64 bytes (8 x 64 bits)
maximum using FPM, your max memory transfer speed will be:

 +----------------------- number of bytes transferred
 |
 |     +----------------- time it takes to read the first word (64 bits)
 |     |
 |     |      +---------- time it takes to read the next 7 words
 |     |      |
 |     |      |      +--- cycle time - row access time
 |     |      |      |    (penalty for going to next row)
 v     v    vvvvv    v
64 / (60 + 30 * 7 + 60) * 10^9 =~ 194 MB/s

I'm not sure how many bytes the Triton can read using FPM, so the
above is just an example.  Also, the above times are in nanoseconds,
and if your memory bus is running at 66MHz, each of them will have to
be rounded up to the next multiple of 15ns (ok, they already are, but
now you know why you don't want to buy a 60MHz machine).

Note that this is a one-way speed out of the memory, so if you are
copying from memory to memory, you will probably have to halve this
(I'm not sure about the write speed, though).  By the way, the fastest 
memory copy I've seen so far is 87MB/s (or 174MB/s in and out) on a
P5-166 with EDO memory.  (See http://now.cs.berkeley.edu/Td/bcopy.html 
for more.)

Matt Thomas replies:

 * >} 64 bit PCI doubles that maximum to 265MB/s.

Does anyone know about 66MHz PCI?  I've seen it in the standards....

 * The main memory access may be done through a PCI host bridge, but that
 * does not mean that's it done through the PCI.  A system/cpu may have

Yes.  These chips are usually identified as:

chip0 <Intel 82440FX (Natoma) PCI and memory controller> rev 2 on pci0:0:0

and

chip0 <Intel 82439HX (Triton II) PCI cache memory controller> rev 1 on pci0:0:0

and such.

Satoshi

From owner-freebsd-hardware  Tue Feb 25 22:50:18 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id WAA20915
          for hardware-outgoing; Tue, 25 Feb 1997 22:50:18 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA20909
          for <freebsd-hardware@freebsd.org>; Tue, 25 Feb 1997 22:50:14 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id RAA06491; Wed, 26 Feb 1997 17:46:32 +1100
Date: Wed, 26 Feb 1997 17:46:32 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702260646.RAA06491@godzilla.zeta.org.au>
To: asami@vader.cs.berkeley.edu, bde@zeta.org.au, jin@george.lbl.gov,
        mark@quickweb.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: freebsd-hardware@freebsd.org, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>*** The same machine with different FreeBSD ***
>
>% dmesg 
>Copyright (c) 1992-1996 FreeBSD Inc.
>Copyright (c) 1982, 1986, 1989, 1991, 1993
>        The Regents of the University of California.  All rights reserved.
>
>FreeBSD 2.2-970215-GAMMA #0: Wed Feb 19 15:22:41 PST 1997
>    root@adv-pc-1.lbl.gov:/usr/src/sys/compile/MinMax
>Calibrating clock(s) relative to mc146818A clock ... i586 clock: 200455533 Hz, i8254 clock: 1193190 Hz
>CPU: Pentium (200.45-MHz 586-class CPU)
>  Origin = "GenuineIntel"  Id = 0x52c  Stepping=12
>  Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
>real memory  = 67108864 (65536K bytes)
>avail memory = 62623744 (61156K bytes)
>
>% dd if=/dev/zero of=/dev/null bs=1m count=1000
>1000+0 records in
>1000+0 records out
>1048576000 bytes transferred in 7.591523 secs (138124591 bytes/sec)
>0.0u 7.5s 0:07.62 99.4% 71+2841k 0+0io 3pf+0w
>
># this result is better than the standard memory bandwidth, but worse than
>the maximum memory bandwidth.
>
>That is why I said "It is O.S. dependent."

That's what I said too.  I mentioned this benchmark as a quick way to
see if you have the i586-optimized copy routines.  It also happens to
expose many poorly tuned memory systems :-).

>The PCI is a 64-bit wide bus. The maxmimum memory speed you can get from
>this bus is 1000000000 * 8 / 60ns = 133333333 Bytes/sec (no inteleave).
>However, no memory sub-system in PC/UNIX O.S. uses 64-bit memory bandwidth
>currently because of the CPU bus.

The memory bandwidth has very little to do with this.  Triton memory buses
are normally 60 or 66 MHz with burst access timing of x-2-2-2 or x-3-3-3.
A P5/133 can almost (*) sustain 6-2-2-2 timing provided the bursts are to
the same page and writes are done 64 bits at a time.  Recent versions
of FreeBSD have copy routines tuned to avoid switching the page and to
write using the FPU.  This gives a memory bandwidth of
1000000000 * 32 / ((6 + 2 + 2 + 2) * 15) = 177777777 Bytes/sec (4/3 times
the PCI bandwidth).  In practice, the current and 2.2 FreeBSD bzero()
achieves 176725593 Bytes/sec for a block size of 4MB.

(*) Copying through the FPU takes a lot of CPU cycles.  `fistpl' must
be used to write.  It takes 6 cycles, so a burst of 4 of these takes 24
cycles - the same time for a 133 MHz CPU as for a 66 MHz memory system
with a 6-2-2-2 burst cycle.  The CPU can't quite keep up, since it has to
do some loop control instructions too.  The loop control instructions only
cost 2-3 cycles, but there is a larger penalty (about 10 more cycles?)
for not keeping up, so the FreeBSD copy routines unroll the loop a few
times so that there is only a penalty for every 256 bytes.  A P5/166
should not have these problems, but a P5/100 is significantly slower
because it can't keep up within the loop.

>Another tip is even some motherboards are using same PCI shipset, the memory
>performance may vary.
>For example, compare ASUS Triton-{I, II} with three different Intel
>motherboards (EV2, PT-2000, ZAPPA) with Triton-II PCI chipset,
>the memory I/O speed does not have much different;
>however, ASUS motherboard can use 70ns memory chip v.s. Intel motherboards have
>to use 60ns memory chip. Very unbelieveable fact, right? This is hardware issue.

The burst timing is different.  The software can sometimes do better if it
knows the precise timing (e.g., when combining writing with reading, it
best to do the reads as late as possible, but if they are too late then
you lose).  This is more of an issue on slow CPUs.  The FreeBSD routines
are tuned for the timing on the authors' systems.

Bruce

From owner-freebsd-hardware  Wed Feb 26 09:52:36 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id JAA21839
          for hardware-outgoing; Wed, 26 Feb 1997 09:52:36 -0800 (PST)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA21832
          for <freebsd-hardware@FreeBSD.ORG>; Wed, 26 Feb 1997 09:52:30 -0800 (PST)
Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) 
  by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id JAA29439;
  Wed, 26 Feb 1997 09:46:28 -0800 (PST)
Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997))  id 8825644A.006147B8 ; Wed, 26 Feb 1997 09:42:33 -0700
X-Lotus-FromDomain: IW @ IWP
From: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
To: bde@zeta.org.au, asami@vader.cs.berkeley.edu, jin@george.lbl.gov,
        mark@quickweb.com
cc: freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Message-ID: <8825644A.00612942.00@IWND1.infoworld.com>
Date: Wed, 26 Feb 1997 10:21:00 -0700
Subject: Re: Memory speed of P6-200 (256k)
Mime-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk






How about MMX, guys? Is there any way to speed up copies by bursting in and
out of the MMX registers? The FPU?



From owner-freebsd-hardware  Wed Feb 26 14:19:41 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id OAA06863
          for hardware-outgoing; Wed, 26 Feb 1997 14:19:41 -0800 (PST)
Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA06858
          for <freebsd-hardware@FreeBSD.ORG>; Wed, 26 Feb 1997 14:19:38 -0800 (PST)
Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.4/8.7.3) id OAA02909; Wed, 26 Feb 1997 14:18:35 -0800 (PST)
Date: Wed, 26 Feb 1997 14:18:35 -0800 (PST)
Message-Id: <199702262218.OAA02909@vader.cs.berkeley.edu>
To: Brett_Glass_at_POST-IW1@infoworld.com
CC: bde@zeta.org.au, jin@george.lbl.gov, mark@quickweb.com,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
In-reply-to: <8825644A.00612942.00@IWND1.infoworld.com> (Brett_Glass_at_POST-IW1@infoworld.com)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * How about MMX, guys? Is there any way to speed up copies by bursting in and
 * out of the MMX registers? The FPU?

The P5-optimized copyin/out already uses the FPU.  We've verified that 
it doesn't help the 486 and P6 (at least with Orion and Natoma).

I don't know about MMX, but from what Bruce has observed, it may help
lower the cost of load/store (which is pretty tough on CPUs for
floating-point ops).  If someone with an MMX can step forward to be
the guinea pig, I'm sure Bruce will be happy to help. :)

Satoshi

From owner-freebsd-hardware  Wed Feb 26 14:30:19 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id OAA07503
          for hardware-outgoing; Wed, 26 Feb 1997 14:30:19 -0800 (PST)
Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07498
          for <freebsd-hardware@FreeBSD.ORG>; Wed, 26 Feb 1997 14:30:16 -0800 (PST)
Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.7.5/8.6.12) with SMTP id RAA12898; Wed, 26 Feb 1997 17:30:36 -0500 (EST)
Date: Wed, 26 Feb 1997 17:30:36 -0500 (EST)
From: Mark Mayo <mark@quickweb.com>
To: Satoshi Asami <asami@vader.cs.berkeley.edu>
cc: Brett_Glass_at_POST-IW1@infoworld.com, bde@zeta.org.au, jin@george.lbl.gov,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Subject: Re: Memory speed of P6-200 (256k)
In-Reply-To: <199702262218.OAA02909@vader.cs.berkeley.edu>
Message-ID: <Pine.BSF.3.95q.970226172737.12865C-100000@vinyl.quickweb.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Wed, 26 Feb 1997, Satoshi Asami wrote:

>  * How about MMX, guys? Is there any way to speed up copies by bursting in and
>  * out of the MMX registers? The FPU?
> 
> The P5-optimized copyin/out already uses the FPU.  We've verified that 
> it doesn't help the 486 and P6 (at least with Orion and Natoma).
> 
> I don't know about MMX, but from what Bruce has observed, it may help
> lower the cost of load/store (which is pretty tough on CPUs for
> floating-point ops).  If someone with an MMX can step forward to be
> the guinea pig, I'm sure Bruce will be happy to help. :)

The problem is that (AFAIK) there aren't any C compilers iwth MMX support
yet.. Even the big guys (Watcom and Microsoft) only have inline assembler
support. Of course, if this stuff is being coded in assembler, you're on
your way!

I'll be getting a MMX machine at school soon to play with (working on
vector stuff - going to prove that Intel MMX is still slower than the
native vectoring operations of the PPC and Alpha) so I can test any code
that shows up.

-Mark

> 
> Satoshi
> 

----------------------------------------------------------------------------
 Mark Mayo		  				mark@quickweb.com       
 RingZero Comp.  	  		   http://vinyl.quickweb.com/mark 
----------------------------------------------------------------------------
 Nature shows that with the growth of intelligence comes increased capacity 
 for pain, and it is only with the highest degree of intelligence that 
 suffering reaches its supreme point.  -- Arthur Schopenhauer


From owner-freebsd-hardware  Wed Feb 26 14:43:01 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id OAA08006
          for hardware-outgoing; Wed, 26 Feb 1997 14:43:01 -0800 (PST)
Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08001
          for <freebsd-hardware@FreeBSD.ORG>; Wed, 26 Feb 1997 14:42:57 -0800 (PST)
Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.4/8.7.3) id OAA02955; Wed, 26 Feb 1997 14:41:59 -0800 (PST)
Date: Wed, 26 Feb 1997 14:41:59 -0800 (PST)
Message-Id: <199702262241.OAA02955@vader.cs.berkeley.edu>
To: mark@quickweb.com
CC: Brett_Glass_at_POST-IW1@infoworld.com, bde@zeta.org.au, jin@george.lbl.gov,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
In-reply-to: <Pine.BSF.3.95q.970226172737.12865C-100000@vinyl.quickweb.com> (message from Mark Mayo on Wed, 26 Feb 1997 17:30:36 -0500 (EST))
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * The problem is that (AFAIK) there aren't any C compilers iwth MMX support
 * yet.. Even the big guys (Watcom and Microsoft) only have inline assembler
 * support. Of course, if this stuff is being coded in assembler, you're on
 * your way!

The routines are in support.s so they will be coded in assembly
language.  Of course, someone has to add them to gas first. :)

 * I'll be getting a MMX machine at school soon to play with (working on
 * vector stuff - going to prove that Intel MMX is still slower than the
 * native vectoring operations of the PPC and Alpha) so I can test any code
 * that shows up.

Cool!

Satoshi

From owner-freebsd-hardware  Wed Feb 26 16:47:14 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id QAA16512
          for hardware-outgoing; Wed, 26 Feb 1997 16:47:14 -0800 (PST)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16506
          for <freebsd-hardware@FreeBSD.ORG>; Wed, 26 Feb 1997 16:47:08 -0800 (PST)
Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) 
  by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id QAA04676;
  Wed, 26 Feb 1997 16:46:02 -0800 (PST)
Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997))  id 8825644B.00039F92 ; Wed, 26 Feb 1997 16:39:34 -0700
X-Lotus-FromDomain: IW @ IWP
From: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
To: asami@vader.cs.berkeley.edu
cc: bde@zeta.org.au, jin@george.lbl.gov, mark@quickweb.com,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Message-ID: <8825644A.0080332B.00@IWND1.infoworld.com>
Date: Wed, 26 Feb 1997 16:01:00 -0700
Subject: Re: Memory speed of P6-200 (256k)
Mime-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk






I've got an MMX test system here, with about a gig of disk to spare. By all
means, fire away!

--Brett



From owner-freebsd-hardware  Thu Feb 27 01:45:43 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id BAA15693
          for hardware-outgoing; Thu, 27 Feb 1997 01:45:43 -0800 (PST)
Received: from cache.lib.itb.ac.id. (cache.lib.ITB.ac.id [167.205.57.94])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA15676
          for <hardware@freebsd.org>; Thu, 27 Feb 1997 01:44:59 -0800 (PST)
Received: from cyberlib.itb.ac.id. (cyberlib.ITB.ac.id [167.205.57.97]) by cache.lib.itb.ac.id. (8.7.5/8.7.3) with SMTP id PAA03438 for <hardware@freebsd.org>; Thu, 27 Feb 1997 15:18:23 GMT
Received: from CYBERLIB/MAILQ by cyberlib.itb.ac.id. (Mercury 1.11);
    Thu, 13 Feb 97 16:51:53 +0700
Received: from MAILQ by CYBERLIB (Mercury 1.11); Thu, 13 Feb 97 16:51:27 +0700
From: "Edo Yudhistira" <EDO101@cyberlib.itb.ac.id>
Organization:  Computer Center ITB
To: hardware@freebsd.org
Date:          Thu, 13 Feb 1997 16:51:14 +07
MIME-Version:  1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject:       MMX
Priority: normal
X-mailer:     Pegasus Mail v3.1 (R1a)
Message-ID: <136CCBE3F15@cyberlib.itb.ac.id.>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


Greetings,
    I was wondering, for you who have the privilege to test the new 
P5 MMX, what's the voltage of this CPU?

Bye.--- Like a wise man used to say: 'Have no fear, Edo is here...'

From owner-freebsd-hardware  Thu Feb 27 09:55:33 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id JAA09961
          for hardware-outgoing; Thu, 27 Feb 1997 09:55:33 -0800 (PST)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA09950
          for <hardware@freebsd.org>; Thu, 27 Feb 1997 09:55:26 -0800 (PST)
Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) 
  by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id JAA10985;
  Thu, 27 Feb 1997 09:53:08 -0800 (PST)
Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997))  id 8825644B.0061E59F ; Thu, 27 Feb 1997 09:49:17 -0700
X-Lotus-FromDomain: IW @ IWP
From: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
To: EDO101@cyberlib.itb.ac.id, hardware@freebsd.org
Message-ID: <8825644B.00614EBE.00@IWND1.infoworld.com>
Date: Thu, 27 Feb 1997 10:01:00 -0700
Subject: MMX
Mime-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk







New Text Item:  MMX

I think it's 2.9V internally, but may be different at the pins.

Intel's spec sheet would say.

--Brett



From owner-freebsd-hardware  Fri Feb 28 01:42:06 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id BAA14009
          for hardware-outgoing; Fri, 28 Feb 1997 01:42:06 -0800 (PST)
Received: from dfw-ix6.ix.netcom.com (dfw-ix6.ix.netcom.com [206.214.98.6])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA13993
          for <freebsd-hardware@FreeBSD.ORG>; Fri, 28 Feb 1997 01:42:03 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix6.ix.netcom.com (8.8.4/8.8.4)
	  id DAA06442; Fri, 28 Feb 1997 03:40:00 -0600 (CST)
Received: from ala-ca9-57.ix.netcom.com(207.93.143.121) by dfw-ix6.ix.netcom.com via smap (V1.3)
	id sma006337; Fri Feb 28 03:39:30 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id BAA05302; Fri, 28 Feb 1997 01:39:22 -0800 (PST)
Date: Fri, 28 Feb 1997 01:39:22 -0800 (PST)
Message-Id: <199702280939.BAA05302@silvia.HIP.Berkeley.EDU>
To: Brett_Glass_at_POST-IW1@infoworld.com
CC: bde@zeta.org.au, jin@george.lbl.gov, mark@quickweb.com,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
In-reply-to: <8825644A.0080332B.00@IWND1.infoworld.com> (Brett_Glass_at_POST-IW1@infoworld.com)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * I've got an MMX test system here, with about a gig of disk to spare. By all
 * means, fire away!

Thanks, but we need (at least) to have gas fixed to understand MMX
instructions before we can even start coding. ;)

I just checked prep but judging from the timestamp of binutils-2.7's
release (2.6->2.7 patch is dated July of last year), it doesn't seem
like FSF has updated it yet.  Does anyone know if someone (like other
*BSD's?) is working on it?

Satoshi

From owner-freebsd-hardware  Fri Feb 28 02:13:06 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id CAA18551
          for hardware-outgoing; Fri, 28 Feb 1997 02:13:06 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA18545
          for <freebsd-hardware@FreeBSD.ORG>; Fri, 28 Feb 1997 02:13:01 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id VAA07624; Fri, 28 Feb 1997 21:08:05 +1100
Date: Fri, 28 Feb 1997 21:08:05 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702281008.VAA07624@godzilla.zeta.org.au>
To: asami@vader.cs.berkeley.edu, Brett_Glass_at_POST-IW1@infoworld.com
Subject: Re: Memory speed of P6-200 (256k)
Cc: bde@zeta.org.au, freebsd-hardware@FreeBSD.ORG, jin@george.lbl.gov,
        kuku@gilberto.physik.rwth-aachen.de, mark@quickweb.com,
        robsch@robkaos.ruhr.de
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> * I've got an MMX test system here, with about a gig of disk to spare. By all
> * means, fire away!
>
>Thanks, but we need (at least) to have gas fixed to understand MMX
>instructions before we can even start coding. ;)

Just write it hex or octal if you are in a hurry.  We already do this
for `rdmsr' etc.

Bruce

From owner-freebsd-hardware  Fri Feb 28 04:20:59 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id EAA23990
          for hardware-outgoing; Fri, 28 Feb 1997 04:20:59 -0800 (PST)
Received: from dfw-ix6.ix.netcom.com (dfw-ix6.ix.netcom.com [206.214.98.6])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA23983
          for <freebsd-hardware@FreeBSD.ORG>; Fri, 28 Feb 1997 04:20:51 -0800 (PST)
Received: (from smap@localhost)
          by dfw-ix6.ix.netcom.com (8.8.4/8.8.4)
	  id GAA22272; Fri, 28 Feb 1997 06:18:59 -0600 (CST)
Received: from ala-ca9-57.ix.netcom.com(207.93.143.121) by dfw-ix6.ix.netcom.com via smap (V1.3)
	id sma021343; Fri Feb 28 06:17:22 1997
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id EAA07192; Fri, 28 Feb 1997 04:17:18 -0800 (PST)
Date: Fri, 28 Feb 1997 04:17:18 -0800 (PST)
Message-Id: <199702281217.EAA07192@silvia.HIP.Berkeley.EDU>
To: bde@zeta.org.au
CC: freebsd-hardware@FreeBSD.ORG
In-reply-to: <199702281008.VAA07624@godzilla.zeta.org.au> (message from Bruce Evans on Fri, 28 Feb 1997 21:08:05 +1100)
Subject: Re: Memory speed of P6-200 (256k)
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * Just write it hex or octal if you are in a hurry.  We already do this
 * for `rdmsr' etc.

Well ok, let's try it first.  However, if what Bruce said is true (and
it usually is) and I understand it correctly (which I usually don't),
we won't get any better results because (1) the memory bandwidth is
about 180MB/s on Triton-I/II with EDO, and (2) a P5-133 can't quite
keep up (I've seen about 160MB/s), while a P6-166 can.  Since MMX
comes only with 166MHz and 200MHz parts, it's not really going to
help.

Anyway.

The following shar contains three files in it: "runtest" (shell
script), "mem.c" (test driver) and "unrolled.s" (assembly language
subroutine).

You can do a

  gcc -O -o mem mem.c unrolled.s

and then

  sh runtest mem

to do the standard FPcopy test.  This should give you around 80MB/s
for a 133MHz and faster Pentium.

Now, look at this part inside "unrolled.s":

===
unrolled_loop:
	fildq 0(%esi)
	fildq 8(%esi)
	fildq 16(%esi)
	fildq 24(%esi)
	fildq 32(%esi)
	fildq 40(%esi)
	fildq 48(%esi)
	fildq 56(%esi)
	fistpq 56(%edi)
	fistpq 48(%edi)
	fistpq 40(%edi)
	fistpq 32(%edi)
	fistpq 24(%edi)
	fistpq 16(%edi)
	fistpq 8(%edi)
	fistpq 0(%edi)
===

These should be changed to

===
unrolled_loop:
	movq mm0,0(%esi)
	movq mm1,8(%esi)
	movq mm2,16(%esi)
	movq mm3,24(%esi)
	movq mm4,32(%esi)
	movq mm5,40(%esi)
	movq mm6,48(%esi)
	movq mm7,56(%esi)
	movq 0(%esi),mm0
	movq 8(%esi),mm1
	movq 16(%esi),mm2
	movq 24(%esi),mm3
	movq 32(%esi),mm4
	movq 40(%esi),mm5
	movq 48(%esi),mm6
	movq 56(%esi),mm7
===

Also, at the end of the routine:

===
	leal -8(%ebp),%esp
	popl %esi
	popl %edi
	leave
	ret
===

you need to insert a "empty mmx state" instruction:

===
	leal -8(%ebp),%esp
	popl %esi
	popl %edi
	.byte 0x0f, 0x77	; emms
	leave
	ret
===

Make these changes and let us know what you get.

Unfortunately, we don't have an assembler that works on this yet so
I'm not sure how to translate the "movq" instructions.  Intel's manual
(that I just downloaded from "http://developer.intel.com/design/mmx/")
says that the opcode for "movq mm, mm/m64" is "0f 6f /r" and "movq
mm/m64 mm" is "0f 7f /r".  Someone who's more proficient in Intel's
terminology can probably translate this.

Satoshi
-------
# This is a shell archive.  Save it in a file, remove anything before
# this line, and then unpack it by entering "sh file".  Note, it may
# create directories; files and directories will be owned by you and
# have default permissions.
#
# This archive contains:
#
#	runtest
#	mem.c
#	unrolled.s
#
echo x - runtest
sed 's/^X//' >runtest << 'END-of-runtest'
X#!/bin/sh
X
Xif [ $# != 1 ]; then
X  echo "usage: $0 executable"
X  exit
Xfi
Xexec=$1
X
Xi=32
Xecho "    size    bandwidth"
Xwhile [ $i -le 1000000 ]; do
X  ./$exec $i 2>&1 >/dev/null | \
X    awk "{printf(\" %7d  %9s MB/s\n\", $i, \$5)}"
X  i=$(($i * 2))
Xdone
END-of-runtest
echo x - mem.c
sed 's/^X//' >mem.c << 'END-of-mem.c'
X/*
X * mem.c - simple memory copy test
X */
X#include <stdio.h>
X#include <stdlib.h>
X#include <string.h>
X#include <ctype.h>
X#include <sys/time.h>
X
X#define TOTALSIZE 4194304	/* 4MB */
X
Xint main(int argc, char **argv)
X{
X        int     usecs, i, bytes;
X        unsigned char *src, *dst;
X	unsigned long tmp;
X	struct timeval tv, start, stop;
X	int N;
X
X	if (argc != 2) {
X		fprintf(stderr,
X		    "Usage: %s size\n", argv[0]);
X		exit(1);
X	}
X	bytes = atoi(argv[1]);
X	if (toupper(argv[1][strlen(argv[1])-1]) == 'K')
X		bytes *= 1024;
X	if (toupper(argv[1][strlen(argv[1])-1]) == 'M')
X		bytes *= (1024 * 1024);
X	if (bytes > TOTALSIZE) {
X	    fprintf(stderr, "size cannot be more than %d\n", TOTALSIZE);
X	    exit(1);
X	}
X	N = TOTALSIZE/bytes;
X        src = (unsigned char *) malloc(bytes*N + 16384);
X	dst = (unsigned char *) malloc(bytes*N + 16384);
X	if (!src || !dst) {
X		perror("malloc");
X		exit(1);
X	}
X
X	/* align arrays to 8K boundaries */
X	tmp = (unsigned long) src;
X	tmp += 8192;
X	tmp &= ~8191;
X	src = (unsigned char *) tmp;
X	tmp = (unsigned long) dst;
X	tmp += 8192;
X	tmp &= ~8191;
X	dst = (unsigned char *) tmp;
X
X	/* fill in src with random junk */
X	gettimeofday(&tv, NULL);
X	srandom(tv.tv_usec);
X	for (i = 0 ; i < bytes*N ; i++)
X	    src[i] = random();
X	bzero(dst, bytes*N);
X
X	/* ensure both src and dst are not swapped out */
X	bcopy(src, dst, bytes*N);
X
X	/* main loop */
X	gettimeofday(&start, NULL) ;
X	for (i = 0; i < N; ++i) {
X	    unrolled(src+bytes*i, dst+bytes*i, bytes);
X	}
X	gettimeofday(&stop, NULL) ;
X
X	usecs = (stop.tv_sec - start.tv_sec)*1000000 +
X	    (stop.tv_usec - start.tv_usec) ;
X
X	/* make sure everything is copied correctly */
X	for (i = 0 ; i < bytes ; i++)
X	    if (src[i] != dst[i])
X		printf("error: byte %d, should be %02x, is %02x\n",
X		       i, src[i], dst[i]);
X
X	fprintf(stderr, "%d bytes copied at  %f MB/s\n", bytes,
X		(double) bytes / 1024 / 1024 / (usecs / 1000000.0 / N));
X
X	return (0);
X}
END-of-mem.c
echo x - unrolled.s
sed 's/^X//' >unrolled.s << 'END-of-unrolled.s'
X.text
X	.align 2
X	.globl	_unrolled
X	.type	_unrolled,@function
X_unrolled:
X	pushl %ebp
X	movl %esp,%ebp
X	pushl %edi
X	pushl %esi
X	movl 8(%ebp),%esi
X	movl 12(%ebp),%edi
X	movl 16(%ebp),%ecx
X
X	cmpl $63,%ecx
X	jbe unrolled_tail
X
X4:
X	pushl %ecx
X	cmpl $1792,%ecx
X	jbe 2f
X	movl $1792,%ecx
X2:
X	subl %ecx,0(%esp)
X	cmpl $256,%ecx
X	jb 5f
X	pushl %esi
X	pushl %ecx
X	.align 4,0x90
X3:
X	movl 0(%esi),%eax
X	movl 32(%esi),%eax
X	movl 64(%esi),%eax
X	movl 96(%esi),%eax
X	movl 128(%esi),%eax
X	movl 160(%esi),%eax
X	movl 192(%esi),%eax
X	movl 224(%esi),%eax
X	addl $256,%esi
X	subl $256,%ecx
X	cmpl $256,%ecx
X	jae 3b
X	popl %ecx
X	popl %esi
X5:
X	.align 2,0x90
Xunrolled_loop:
X	fildq 0(%esi)
X	fildq 8(%esi)
X	fildq 16(%esi)
X	fildq 24(%esi)
X	fildq 32(%esi)
X	fildq 40(%esi)
X	fildq 48(%esi)
X	fildq 56(%esi)
X	fistpq 56(%edi)
X	fistpq 48(%edi)
X	fistpq 40(%edi)
X	fistpq 32(%edi)
X	fistpq 24(%edi)
X	fistpq 16(%edi)
X	fistpq 8(%edi)
X	fistpq 0(%edi)
X	addl $-64,%ecx
X	addl $64,%esi
X	addl $64,%edi
X	cmpl $63,%ecx
X	ja unrolled_loop
X	popl %eax
X	addl %eax,%ecx
X	cmpl $64,%ecx
X	jae 4b
X
Xunrolled_tail:
X	movl %ecx,%eax
X	shrl $2,%ecx
X	cld
X	rep
X	movsl
X	movl %eax,%ecx
X	andl $3,%ecx
X	rep
X	movsb
X
X	leal -8(%ebp),%esp
X	popl %esi
X	popl %edi
X	leave
X	ret
X
Xunrolled_end:
X	.size	_unrolled,unrolled_end-_unrolled
END-of-unrolled.s
exit


From owner-freebsd-hardware  Fri Feb 28 10:56:15 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id KAA11265
          for hardware-outgoing; Fri, 28 Feb 1997 10:56:15 -0800 (PST)
Received: from lserver.infoworld.com (lserver.infoworld.com [192.216.48.4])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA11259
          for <freebsd-hardware@FreeBSD.ORG>; Fri, 28 Feb 1997 10:56:12 -0800 (PST)
Received: from IWND1.infoworld.com (iwnd1.infoworld.com [192.216.49.131]) 
  by lserver.infoworld.com (8.8.5/8.8.5/GNAC-GW-2.1) with SMTP id KAA23820;
  Fri, 28 Feb 1997 10:55:18 -0800 (PST)
Received: by IWND1.infoworld.com(Lotus SMTP MTA v1.05 (305.3 1-15-1997))  id 8825644C.0067958D ; Fri, 28 Feb 1997 10:51:24 -0700
X-Lotus-FromDomain: IW @ IWP
From: "Brett Glass at POST-IW1"<Brett_Glass_at_POST-IW1@infoworld.com>
To: asami@vader.cs.berkeley.edu
cc: bde@zeta.org.au, jin@george.lbl.gov, mark@quickweb.com,
        freebsd-hardware@FreeBSD.ORG, kuku@gilberto.physik.rwth-aachen.de,
        robsch@robkaos.ruhr.de
Message-ID: <8825644C.00677745.00@IWND1.infoworld.com>
Date: Fri, 28 Feb 1997 11:42:00 -0700
Subject: Re: Memory speed of P6-200 (256k)
Mime-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk







New Text Item:  Re: Memory speed of P6-200 (256k)

>  Thanks, but we need (at least) to have gas fixed to understand MMX
>  instructions before we can even start coding. ;)

Why? Just make the instructions into macros that emit bytes.

--Brett



From owner-freebsd-hardware  Sat Mar  1 09:47:06 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id JAA21898
          for hardware-outgoing; Sat, 1 Mar 1997 09:47:06 -0800 (PST)
Received: from arrakis.cs.put.poznan.pl (root@arrakis.cs.put.poznan.pl [150.254.33.194])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA21886
          for <freebsd-hardware@FreeBSD.ORG>; Sat, 1 Mar 1997 09:47:00 -0800 (PST)
Received: (from piesik@localhost)
          by arrakis.cs.put.poznan.pl (8.8.5/8.8.4)
	  id SAA05709 for freebsd-hardware@FreeBSD.ORG; Sat, 1 Mar 1997 18:46:56 +0100 (MET)
Date: Sat, 1 Mar 1997 18:46:56 +0100 (MET)
From: Piotr Piesik <piesik@arrakis.cs.put.poznan.pl>
Message-Id: <199703011746.SAA05709@arrakis.cs.put.poznan.pl>
To: freebsd-hardware@FreeBSD.ORG
Subject: DPT SCSI
Sender: owner-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk


hi :-)

Is it possible to use DPT SCSI controller with FreeBSD?
I'm getting message (during boot) like this:

pci0:18:DPT,device=0xa400,class=storage(scsi) int a irq 9 (no driver assigned)

greetings

Piotr Piesik
Poznan University of Technology

From owner-freebsd-hardware  Sat Mar  1 13:29:10 1997
Return-Path: <owner-hardware>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id NAA00325
          for hardware-outgoing; Sat, 1 Mar 1997 13:29:10 -0800 (PST)
Received: from execulink.com (execulink.com [207.216.160.10])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA00320
          for <freebsd-hardware@freebsd.org>; Sat, 1 Mar 1997 13:29:05 -0800 (PST)
Received: from empey (PC-531.on.rogers.wave.ca [24.112.48.46]) by execulink.com (8.8.4/8.7.3) with ESMTP id QAA13237 for <freebsd-hardware@freebsd.org>; Sat, 1 Mar 1997 16:28:32 -0500 (EST)
Received: from Spooler by empey (Mercury/32 1.22);
    1 Mar 97 16:28:30 +0500
Received: from spooler by integral.on.ca (Mercury/32 1.23); 1 Mar 97 16:27:49 +0500
From: "David Empey" <empey@integral.on.ca>
Organization: Integral Communications
To: freebsd-hardware@freebsd.org
Date: Sat, 1 Mar 1997 16:27:37 +0500
MIME-Version: 1.0
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Subject: Network Card - SMC 8416
Reply-to: empey@integral.on.ca
Priority: normal
X-mailer: Pegasus Mail for Windows (v2.42a)
Message-ID: <8E59B0082@integral.on.ca>
Sender: owner-hardware@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

I have a network card installed, an SMC 8416, which doesn't appear to
be supported .  I tried the driver for the SMC 80XX cards to no avail.
Any help appreciated!

___________________________________________________________

David Empey
mailto: empey@integral.on.ca
homepage:  http://www.integral.on.ca/empey/ 
604-185 Berkshire Drive, London, Ontario, Canada, N6J 3R6
(519)-474-0296