Messages in this thread | | | Date | Mon, 10 Nov 1997 12:55:44 +0000 | From | (Neilski) | Subject | Re: Pentium DEATH and Cyrix COMA in user-mode |
| |
Andrew Kieschnick wrote: >I still say that they're also significantly different in that the pentium >is executing what should be an invalid instruction, and the cyrix is >executing valid instructions stupidly.
Totally agree, BUT, don't forget that (AFAIK) people found the Cyrix bug by poking about after the Pentium bug was found (and indeed these valid instructions break it), so has anyone tried the Cyrix-breaking opcodes on an Intel Pentium ? Could be that the Pentium has other problems relating to lock prefixes that cause it to hang even with some *valid* opcodes.
Neil
| |