trusting the processor chip

[Rick Smith]

cwe at it.kth.se (Christian Wettergren) writes:

>Take a look at the IEEE Symp on Security and Privacy Proceedings from
>1995, I believe it was. There was a paper there about security bugs in
>the Intel processors, enumerating a number of them in 80386 for example.
>There where at least one or two byte sequences that plainly stopped 
>the processor.

Yes, and this is where the real risks are. The original question was
entirely about explicit subversion. The larger risk is accidental
flaws. Same with software in most cases.

Rick.
smith at sctc.com           secure computing corporation