Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Petz who wrote (25752)	11/10/1997 2:53:00 AM
From: Paul Engel	Read Replies (1) \| Respond to of 1579914

Petz - Re: " Who's to say this bit pattern isn't the middle bytes of some floating point number sitting in memory that accidentally gets jumped to by a buggy program" Conclusion: Buggy Code is OK. Paul

To: Petz who wrote (25752)	11/10/1997 11:37:00 AM
From: Ali Chen	Respond to of 1579914

Petz: <is this bad [Pentium] opcode 16 bits wide?> No, it is actually an invalid form of rare 64-bit instruction: "Compare and Exchange 8 Bytes", with LOCK prefix, LOCK CMPXCHG8B AX. The destination AX is already used in the instruction as a part of the source, so the conflict is apparent. However, the Pentium manual claims to set #UD exception in this case, but you know the result - deep OS crash. Of course, it is possible to run into this combination ocassionally with all these buggy programs, but who cares to calculate the probability? The point now is that the recipe is known for malicious people, and the threat for small ISP is real. Ali