From ... From: Erik Naggum Subject: Re: Evidence of an insidious Pentium Bug? Date: 1999/09/03 Message-ID: <3145341043550221@naggum.no>#1/1 X-Deja-AN: 520548728 References: mail-copies-to: never X-Complaints-To: usenet@news.eunet.no X-Trace: oslo-nntp.eunet.no 936352244 10353 193.71.66.49 (3 Sep 1999 09:50:44 GMT) Organization: Naggum Software; +47 8800 8879; +1 510 435 8604; http://www.naggum.no NNTP-Posting-Date: 3 Sep 1999 09:50:44 GMT Newsgroups: comp.lang.lisp * "David McClain" | The code snippet below compiles to a harmless and correct sequence of | Pentium instructions, yet the version that negates by subtraction from | zero works correctly while the direct negation with "FCHS" fails, in the | routines "catanh" and "catanhx". why post this in ML when you're really trying to show a problem in the Pentium? I'd appreciate the assembly listing and discussion about that. | My confidence in the Pentium architecture has now been quite badly shaken! wow, it took you this long? ;) #:Erik -- save the children: just say NO to sex with pro-lifers