Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 14 Oct 1996 19:11:02 +0500
From:      A JOSEPH KOSHY <koshy@india.hp.com>
To:        Joe Greco <jgreco@brasil.moneng.mei.com>
Cc:        hackers@freebsd.org, hardware@freebsd.org
Subject:   Re: AMD 586 runs FreeBSD just FINE 
Message-ID:  <199610141411.AA107472262@fakir.india.hp.com>
In-Reply-To: Your message of "Mon, 14 Oct 1996 08:05:18 EST." <199610141305.IAA23662@brasil.moneng.mei.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
>>>> "Joe Greco" <jgreco@brasil.moneng.mei.com> writes

> That's odd (I think)...  I upgraded an ASUS SP3G board from a DX2/66
> to a DX5/133 without a single jumper change...  from the SP3G's point
> of view it was a _real_ fast DX2/66  :-)

I've been running an Am5x86 for some months now on a SiS motherboard.

Replacing CPU's isn't generally as simply as dropping in a new CPU
over the old one.  For example:

a) The Am5x86 supports a write-back L1 cache; since most 486's offer a 
   write-thru L1 cache your motherboard needs to understand enough of the
   AMD bus protocol to allow the chip to work correctly in the presence of
   other bus masters etc.

b) Further, since the chip is overclocked wrt to the system bus, the 
   motherboard needs to feed the correct clock frequencies to the chip.

c) Then there is the matter of supply voltages; some of the newer chips
   run off 3.3 and 3.45 Volts while the older 486's ran off 5V.  The 
   motherboard needs to know the right voltage levels to feed the CPU.

d) Finally there are pins on some of the newer chips that have defined
   functionality now which were earlier N/Cs.  An example would be the
   extra pins needed to support cache snooping with a write-back cache
   in the CPU.  The motherboard needs to drive these correctly.

Thus some form of jumpering is the norm when changing CPUs.  Esp. in
case (c) above.

Koshy



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610141411.AA107472262>