Security question (sort of)

From: Jason Willgruber <roblwill_at_usaor.net>
Date: Sat Mar 20 21:18:07 1999

If the ART chips that drive the serial ports are gone, they won't work no
matter what it used to access them.

--
         -Jason Willgruber
       (roblwill_at_usaor.net)
           ICQ#: 1730318
<http://members.tripod.com/general_1>
------/////////
-----//     //
----//       ##
---//   ////###
--//     //####
-/////////## ##
-----------/ ##
          /  ##
         /   ##
        / ########
       /----------->
-----Original Message-----
From: Eric Smith <eric_at_brouhaha.com>
To: Discussion re-collecting of classic computers
<classiccmp_at_u.washington.edu>
Date: Friday, March 19, 1999 10:25 PM
Subject: Re: Security question (sort of)
>Jason wrote:
>
>> A virus planted by a hacker can damage hardware by "eating" at the chips,
>
>I'm not a virus expert, but I am an experienced embedded systems
programmer,
>and have done some hardware design.  That said, this claim sounds
completely
>ridiculous to me, on par with an urban legend.  Care to explain this from a
>electronics (or physics, or chemistry) point of view?
>
>> or just scrambling the code in the chip.
>
>In a flash BIOS, maybe, although that would be tend to be specific to a
>certain motherboard.  There's no general way to write a virus that can
trash
>the BIOS on any arbitrary motherboard, because unlike much of the
>"PC standard", there is not a stanard for how the flash BIOS programming
>works.  Different motherboards use different types of flash chips that
>have different programming requirements.
>
>I've never yet heard of a virus doing this, although I'll concede that it
is
>possible.  Decent motherboards require you to physically move a jumper in
>order to enable programming the BIOS, to prevent exactly this kind of
problem.
>
>However, changing the BIOS such that the machine still booted but simply
>didn't have INT 13 serial support wouldn't prevent all software from
>using the serial port.  Most software these days doesn't even bother to
>go through the BIOS to access the ports, because (1) the BIOS interface
>is incredibly lame, and (2) on some machines the BIOS functions don't even
>work correctly.
>
>> (I know someone (Ironically, it's the sister of the person that did this
>> to my computer), who's keyboard controller chip got scrambled.
>
>Sounds like a complete coincidence to me.  Except for exotic (and fairly
>expensive) keyboards, the firmware is in masked ROM inside a
microcontroller,
>and there is no way to modify it without physically replacing the chip.
>
>Eric
>

Received on Sat Mar 20 1999 - 21:18:07 GMT