[5513] in bugtraq
Re: Intel Pentium Bug
daemon@ATHENA.MIT.EDU (Stefan Hudson)
Fri Nov 7 23:29:15 1997
Date: Fri, 7 Nov 1997 19:49:17 -0800
Reply-To: Stefan Hudson <hudson@MBAY.NET>
From: Stefan Hudson <hudson@MBAY.NET>
X-To: "Kragen \"Skewed\" Sitaker" <kragen@DNACO.NET>
To: BUGTRAQ@NETSPACE.ORG
In-Reply-To: <Pine.SUN.3.96.971107193420.14978A-100000@picard.dnaco.net>; from
Kragen "Skewed" Sitaker on Fri, Nov 07, 1997 at 07:35:48PM -0500
On Fri, Nov 07, 1997 at 07:35:48PM -0500, Kragen "Skewed" Sitaker wrote:
> If Intel releases code to update the microcode, someone else will figure
> out how to do it, too. That means we can look forward to a future of
> microcode back-doors and viruses. (How big is the microcode on a Pentium,
> I wonder?)
I'd imagine that there would be some sort of hardware access to the
chip required to do this; for example, putting the chip in a special
motherboard that puts programming voltage on a certain pin. Laptops
with integrated processors might have a jumper or solder pad that
could be used to apply the voltage. A special instruction would then
be used to put the chip in programming mode and upload the new code.
A higher than normal voltage would be required anyway if the code was
stored in EEPROM type memory, I believe.
This would make the microcode patch a lot more complicated, but it's
still better than shipping it back to Intel for replacement, and it
eliminates any possibility of "unintended" microcode updates. They
said it was upgradable; I don't think they ever said it was
SOFTWARE upgradable.
--
/// Stefan Hudson <hudson@mbay.net>
__ /// Senior Network Administrator - Monterey Bay Internet
\\\/// http://www.mbay.net/ - Email: info@mbay.net
\XX/ Voice: 408-642-6100 Fax: 408-642-6101 Modem: 408-642-6102
