Path: utzoo!attcan!uunet!decwrl!wuarchive!mit-eddie!uw-beaver!rice!titan.rice.edu!preston From: preston@titan.rice.edu (Preston Briggs) Newsgroups: comp.arch Subject: Re: speculative execution Message-ID: <1990Oct9.162639.23516@rice.edu> Date: 9 Oct 90 16:26:39 GMT References: Sender: news@rice.edu (News) Organization: Rice University, Houston Lines: 17 In article mh2f+@andrew.cmu.edu (Mark Hahn) writes: >can anyone say anything about architectures for speculative execution? >I think the Microprocessor Report mentioned that a new sparc >implementation will do it. isn't the 88000 (over-)due for a major >overhaul, too? Well sure. Compilers can schedule instructions to be executed speculatively (or optimistically) on all sorts of machines. Wide instruction words offer opportunites, as do long pipelines. You could probably do it in hardware too (that it, go shooting off down some code path before knowing a condition), but I expect compilers can do a better job. Think in terms of trace scheduling. -- Preston Briggs looking for the great leap forward preston@titan.rice.edu