home *** CD-ROM | disk | FTP | other *** search
Wrap
Received: from yonge.cs.toronto.edu (yonge.cs.toronto.edu [128.100.1.8]) by nacm.com (8.6.9/8.6.9) with SMTP id UAA02823 for <executor@nacm.com>; Fri, 4 Nov 1994 20:59:39 -0800 Received: from neat.cs.toronto.edu ([128.100.1.65]) by yonge.cs.toronto.edu with SMTP id <62532>; Fri, 4 Nov 1994 23:59:33 -0500 Received: by neat.cs.toronto.edu id <6164>; Fri, 4 Nov 1994 23:59:17 -0500 From: Jeff Tupper <mooncake@cs.toronto.edu> To: executor@nacm.com Subject: Re: How to generate discussion Message-Id: <94Nov4.235917est.6164@neat.cs.toronto.edu> Date: Fri, 4 Nov 1994 23:59:16 -0500 Sender: Executor-Owner@nacm.com Precedence: bulk Well, thank you for the reply. Does executor store the native code it generates on disk or does it regenerate on every restart of the program? I was looking at the table comparing performance of the Quadra 610 running 68K code on its 68040 vs the synthetic CPU on various IBM boxes... would it be possible to compare vs IBM code on IBM boxes? It seems to me that such a comparison (say 486/66 running the synthetic benchmarks vs IBM code on same benchmarks) would show how much farther the synthetic CPU has to go before matching the performance of compiled code for the IBM. In theory, it seems to me that the synthetic CPU should be able to match the real one (it would take alot of analysis... data flow and control flow analysis just like in a real compiler/optimizer). Am I way off base here? Or is such analysis too complicated for now? Are there reaons besides time constraints that heavy optimization isn't performed and then the results stored on disk? One more question... how many people are on this list? Also, I should state that I am glad you guys are up to what you are up to :)