From drew@hazelrah.cs.Colorado.EDU  Ukn Jul  8 05:07:48 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA15478; Thu, 8 Jul 93 05:07:47 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA11205; Thu, 8 Jul 93 02:07:04 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA07256; Thu, 8 Jul 93 02:06:46 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA07246; Thu, 8 Jul 93 02:06:43 -0700
Received: from hazelrah.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA11201; Thu, 8 Jul 93 02:06:54 -0700
Received: from localhost by hazelrah.cs.Colorado.EDU with SMTP id AA28443
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Thu, 8 Jul 1993 03:06:34 -0600
Message-Id: <199307080906.AA28443@hazelrah.cs.Colorado.EDU>
To: riscy@pyramid.com
Subject: MIPS R4200 / IDT3081 question
Date: Thu, 08 Jul 1993 03:06:32 -0600
From: Drew Eckhardt <drew@hazelrah.cs.Colorado.EDU>
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 


What would having sockets for the 3081 and 4200 take 
(provided that people aren't stupid enough to plug  
chips in both sockets at the same time)?

This would give us a working board *soon* so that Linux 
can get ported and stable, and would allow an upgrade 
path to the 4200 when MIPS finally ships it a year after
the announced date :-)


 
From hasse@inf.ethz.ch  Ukn Jul  8 05:29:37 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA15790; Thu, 8 Jul 93 05:29:35 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA12838; Thu, 8 Jul 93 02:27:02 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08870; Thu, 8 Jul 93 02:26:43 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA08860; Thu, 8 Jul 93 02:26:41 -0700
Received: from bernina.ethz.ch 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA12834; Thu, 8 Jul 93 02:26:53 -0700
Received: from neptune by bernina.ethz.ch with SMTP inbound id <24533-0@bernina.ethz.ch>; Thu, 8 Jul 1993 11:26:19 +0200
Message-Id: <9307080926.AA10491@neptune>
Received: from apelc1.inf.ethz.ch by neptune id AA10491; Thu, 8 Jul 93 11:26:13 +0200
To: riscy@pyramid.com
Cc: hasse@inf.ethz.ch
Subject: Re: MIPS R4200 / IDT3081 question 
In-Reply-To: Your message of "Thu, 08 Jul 93 03:06:32 MDT."
             <199307080906.AA28443@hazelrah.cs.Colorado.EDU> 
Date: Thu, 08 Jul 93 11:25:56 +0200
From: hasse@inf.ethz.ch
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 

Until now I just listen what others said. Now Drew Eckhardt
suggested two sockets for CPUs. I like this idea but in a slightly
different way. Since I'm not an ee-guy I don't know about technical
issues. But what about a two processor system and who ever likes to
have more cpu-power plugs in another cpu. I like the responsiveness
(is this an english word) of multi-processor machines. Everything
runs much smoother. I'm not 100% aware of the impacts to Linux.
I know the best solution will be hard (multi-threaded kernel).
BTW, this will give us in case of a dump frame buffer (didn't we
have the idea of a second cpu for the video before) the scalability
depending on the money one wants to spend for the whole system.

  CH

 
From drew@alta.cs.Colorado.EDU  Ukn Jul  8 05:56:31 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA16207; Thu, 8 Jul 93 05:56:30 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15180; Thu, 8 Jul 93 02:54:20 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA11954; Thu, 8 Jul 93 02:54:02 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA11944; Thu, 8 Jul 93 02:53:59 -0700
Received: from alta.cs.colorado.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA15176; Thu, 8 Jul 93 02:54:10 -0700
Received: from localhost by alta.cs.Colorado.EDU with SMTP id AA27473
  (5.65c/IDA-1.4.4 for <riscy@pyramid.com>); Thu, 8 Jul 1993 03:53:47 -0600
Message-Id: <199307080953.AA27473@alta.cs.Colorado.EDU>
To: riscy@pyramid.com
Cc: hasse@inf.ethz.ch
Subject: Re: MIPS R4200 / IDT3081 question 
In-Reply-To: Your message of "Thu, 08 Jul 1993 11:25:56 +0200."
             <9307080926.AA10491@neptune> 
Date: Thu, 08 Jul 1993 03:49:40 +0100
From: Drew Eckhardt <drew@alta.cs.Colorado.EDU>
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 


--------

    Until now I just listen what others said. Now Drew Eckhardt
    suggested two sockets for CPUs. I like this idea but in a slightly
    different way. Since I'm not an ee-guy I don't know about technical
    issues. But what about a two processor system and who ever likes to
    have more cpu-power plugs in another cpu. I like the responsiveness
    (is this an english word) of multi-processor machines. Everything
    runs much smoother. 

Memory bandwidth is the bottleneck on everything but super computers.  
Throw two processors at a problem in a shared memory multi-processor
design and not enough cache (remember, we aren't using a second level
cache because of cost reasons) things won't get any faster.

    I'm not 100% aware of the impacts to Linux.
    I know the best solution will be hard (multi-threaded kernel).

You could *probably* get away with mutexes in appropriate parts 
of the kernel, although it would be ugly.

    BTW, this will give us in case of a dump frame buffer (didn't we
    have the idea of a second cpu for the video before) the scalability
    depending on the money one wants to spend for the whole system.

IMHO, the added design complexity in both hardware and software probably 
isn't worth it, since our goal is "reasonable performance at a reasonable 
cost".

If you aren't satisfied with Linux's response : 

Basically, operating systems just sit arround and allocate 
resources, making tradeoffs between things like fairness, 
performance, and response time.

In Linux, we trade off fairness for interactive response time 
when we put processes at the head of the run queue when they're
woken up rather than the tail.  We trade performance for fairness
when we reserve the last buffer cache requests for reads rather
than writes.  We trade performance (ie, minimal scheduler overhead)
for interactive response.

If you feel that Linux isn't responsive enough, change the priorities
(ie, modify the scheduler, buffer cache code, whatever) to suit your
needs.

 
From andy@resi.waldorf-gmbh.de  Ukn Jul  8 06:13:38 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA16433; Thu, 8 Jul 93 06:13:37 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16377; Thu, 8 Jul 93 03:12:21 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13226; Thu, 8 Jul 93 03:11:56 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA13214; Thu, 8 Jul 93 03:11:52 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA16226; Thu, 8 Jul 93 03:12:02 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.e) via EUnet
	id eA13498; Thu, 8 Jul 1993 12:08:49 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for unido
	id AA13557; Thu, 8 Jul 93 11:33:30 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Thu, 8 Jul 93 11:32:10 +0200
Message-Id: <9307080932.AA14731@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA14731; Thu, 8 Jul 93 11:32:10 +0200
To: riscy@pyramid.com
Subject: Re:  MIPS R4200 / IDT3081 question
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 


> What would having sockets for the 3081 and 4200 take 
> (provided that people aren't stupid enough to plug  
> chips in both sockets at the same time)?

Good idea, although I don't know if this is possible.
I am still waiting for an official statement from IDT.
I expect it today, perhaps tomorrow.

Andy

-------------------------------------------------------------------------------
Waldorf Electronics GmbH, R&D Department
c/o Andreas Busse
Neustrasse 9-12
D-5481 Waldorf
Phone:  +49 (0)2636-80294
Fax:    +49 (0)2636-80188
e-mail: andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------

 
From caret@pyramid.com  Ukn Jul  8 22:14:54 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA26979; Thu, 8 Jul 93 22:14:52 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA19448; Thu, 8 Jul 93 19:14:14 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12355; Thu, 8 Jul 93 19:13:51 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA12344; Thu, 8 Jul 93 19:13:49 -0700
From: caret@pyramid.com (Neil Russell)
Message-Id: <9307090213.AA12344@sword.eng.pyramid.com>
Subject: Re: MIPS R4200 / IDT3081 question
To: riscy@pyramid.com
Date: Thu, 8 Jul 93 19:13:49 PDT
In-Reply-To: <199307080906.AA28443@hazelrah.cs.Colorado.EDU>; from "Drew Eckhardt" at Jul 8, 93 3:06 am
X-Mailer: ELM [version 2.3 PL11]
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 

> What would having sockets for the 3081 and 4200 take 
> (provided that people aren't stupid enough to plug  
> chips in both sockets at the same time)?
> 
> This would give us a working board *soon* so that Linux 
> can get ported and stable, and would allow an upgrade 
> path to the 4200 when MIPS finally ships it a year after
> the announced date :-)

Please don't forget that to design a board that has the option of
installing a R4200 (or any other part/subsystem), then that must be
tested.  To test it requires a chip!  In any case, who knows what the
bus interface is like on this chip.  If its much different from the IDT
(likely) then we have to have lots of logic to interface it, and thus,
lots of board space.  If we want to go with the R4?00, then we should
wait until it gets released, then start all of this again.

-- 
Neil Russell		(The wizard from OZ)
Pyramid Technology			Email:  caret@pyramid.com
3860 N. First Street			Voice:  (408) 428-7302
San Jose, CA 95134-1702			  FAX:  (408) 428-8845

 
From andy@resi.waldorf-gmbh.de  Ukn Jul  9 03:07:15 1993
Received: from gossip.pyramid.com by SunSITE.unc.edu (4.1/tas-gen/1-30-93)
	id AA04433; Fri, 9 Jul 93 03:07:12 EDT
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA23118; Fri, 9 Jul 93 00:05:36 -0700
Received: by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09319; Fri, 9 Jul 93 00:05:06 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA09301; Fri, 9 Jul 93 00:05:04 -0700
Received: from mail.Germany.EU.net 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA23108; Fri, 9 Jul 93 00:05:14 -0700
Received: by mail.Germany.EU.net(EUnetD-2.3.0.e) via EUnet
	id fD27440; Fri, 9 Jul 1993 09:02:00 +0200
Received: from resi 
	by wegy.waldorf-gmbh.de with SMTP (5.61/GEN-1.0.7)
	via EUnet for unido
	id AA20142; Fri, 9 Jul 93 08:58:21 +0200
From: Andreas Busse <andy@resi.waldorf-gmbh.de>
Date: Fri, 9 Jul 93 08:57:00 +0200
Message-Id: <9307090657.AA16341@resi.waldorf-gmbh.de>
Received: by resi.waldorf-gmbh.de (5.61/GEN-1.0.7)
	via EUnet for wegy.waldorf-gmbh.de
	id AA16341; Fri, 9 Jul 93 08:57:00 +0200
To: riscy@pyramid.com
Subject: Re: MIPS R4200 / IDT3081 question
Sender: riscy-request@pyramid.com
Reply-To: riscy@pyramid.com
Status: RO
X-Status: 


> > This would give us a working board *soon* so that Linux 
> > can get ported and stable, and would allow an upgrade 
> > path to the 4200 when MIPS finally ships it a year after
> > the announced date :-)

> Please don't forget that to design a board that has the option of
> installing a R4200 (or any other part/subsystem), then that must be
> tested.  To test it requires a chip!  In any case, who knows what the
> bus interface is like on this chip.  If its much different from the IDT
> (likely) then we have to have lots of logic to interface it, and thus,
> lots of board space.  If we want to go with the R4?00, then we should
> wait until it gets released, then start all of this again.

I totally agree. If we don't get all necessary information about
the R4K2 within a few days and samples within a few weeks, we should
forget about that. And this is very unlikely.

Andy

-------------------------------------------------------------------------------
Waldorf Electronics GmbH, R&D Department
c/o Andreas Busse
Neustrasse 9-12
D-5481 Waldorf
Phone:  +49 (0)2636-80294
Fax:    +49 (0)2636-80188
e-mail: andy@resi.waldorf-gmbh.de
-------------------------------------------------------------------------------

 
