Re: pentium bug continued

Dale Amon as Operator (root@starbase1.gpl.net)
Tue, 11 Nov 97 18:37:19 GMT


> Except that in this case the instruction does not behave according to
> Intel's own Pentium documentation (241430_4.pdf, page 25-71, there may be
> a more recent version):

But whether the problem opcode is an error in a real opcode, an undocumented
test opcode or an accidental opcode, the end result is the same - a cpu
that is locked up hard. This is not at all a nice thing to have floating
around. It's also perhaps a good reason for lower complexity, better
tested instruction sets, but that is a set of PhD theses best left alone.
We're rather stuck with the Intel set (which I've despised since the
8008)

But back to the thread - I think we're rather stuck insofar as the
average machine is concerned. This could have serious impacts on
machines used for customer shell accounts. If I start seeing any
unexplained crashes I'll be the first to let all of you know.