From owner-freebsd-hardware  Tue Aug 12 20:57:37 1997
Return-Path: <owner-freebsd-hardware>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id UAA17917
          for hardware-outgoing; Tue, 12 Aug 1997 20:57:37 -0700 (PDT)
Received: from www2.shoppersnet.com (shoppersnet.com [204.156.152.112])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA17912
          for <hardware@FreeBSD.ORG>; Tue, 12 Aug 1997 20:57:32 -0700 (PDT)
Received: (from hlew@localhost)
	by www2.shoppersnet.com (8.8.5/8.8.5) id VAA23419;
	Tue, 12 Aug 1997 21:02:16 -0700 (PDT)
Date: Tue, 12 Aug 1997 21:02:15 -0700 (PDT)
From: Howard Lew <hlew@www2.shoppersnet.com>
To: Mike Burgett <mburgett@cmnsens.zoom.com>
cc: "hardware@freebsd.org" <hardware@FreeBSD.ORG>
Subject: Re: More K6 info...
In-Reply-To: <199708120511.WAA07110@dragon.cmnsens.zoom.com>
Message-ID: <Pine.BSF.3.91.970812204118.22554A-100000@www2.shoppersnet.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hardware@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Mon, 11 Aug 1997, Mike Burgett wrote:

> I don't know what insight might be gleaned from it, but I've seen the 
> following types of errors in my numereous attempts to build:
> 
> I've seen (I think) at least these different failures:
> 
> signal 10 or 11 in cc, make or as
> 
> Also, I've seen it choke on code in a .c, .s or temporary compiler file in  
> /var/tmp, where it looks like it 'misread' the file (i.e., the error was 
> bogus...)

Hmmm... I have done some reading in the chips newsgroup and someone
mentioned that running the K6-233 had a real cool heatsink when running
NT4 but was burning hot with Win95.  There is talk that this maybe because
NT uses the "halt" instruction when the cpu idles.  This may explain why
some people with FreeBSD have said that make world fails after about 1.5
hours or so. 

I guess if the issue is heat-related then there are several options:

1) Use a Peltier cooler on the K6 if you need to keep it working hard all 
the time.  Would a Cyrix 6x86 fan do?

2) Implement a "halt" like feature in the kernel configuration for the K6 
(don't know if this is possible though)

3) Run the K6 at a lower speed (not a popular idea of course and won't 
work for 166 users)... 

Perhaps the new 233-300 will be better... According to FreeTech's web
site, the new K6-233, 266, and 300s will use 4.15, 4.75, and 5.40 Amps 
and run at 2.00V so they should run much cooler.  This is in contrast to 
the 6.25, 7.5, and 9.5 Amps that the 166, 200, and 233 require now.

http://www.freetech.com/INFO/97T/97t02.htm