From bill@cc.gatech.edu Wed Nov  6 17:44:37 1996
Received: from burdell.cc.gatech.edu (root@burdell [130.207.3.207]) by cleon.cc.gatech.edu (8.8.Beta.5/8.6.9) with ESMTP id RAA02710 for <annc@cleon.cc.gatech.edu>; Wed, 6 Nov 1996 17:44:35 -0500 (EST)
Received: from forge.cc.gatech.edu (bill@forge.cc.gatech.edu [130.207.9.19]) by burdell.cc.gatech.edu (8.8.Beta.5/8.6.9) with ESMTP id RAA27145 for <annc@cc.gatech.edu>; Wed, 6 Nov 1996 17:44:34 -0500 (EST)
Received: (from bill@localhost) by forge.cc.gatech.edu (8.8.Beta.5/8.6.9) id RAA11346 for annc@cc; Wed, 6 Nov 1996 17:41:48 -0500 (EST)
Date: Wed, 6 Nov 1996 17:41:48 -0500 (EST)
From: bill@cc.gatech.edu (Bill Appelbe)
Message-Id: <199611062241.RAA11346@forge.cc.gatech.edu>
To: annc@cc.gatech.edu
Subject: benchmark status ...
Status: R

	From reid@cc.gatech.edu Tue Nov  5 22:08:10 1996
	Received: from burdell.cc.gatech.edu (root@burdell [130.207.3.207]) by forge.cc.gatech.edu (8.8.Beta.5/8.6.9) with ESMTP id WAA29409 for <bill@forge.cc.gatech.edu>; Tue, 5 Nov 1996 22:08:09 -0500 (EST)
	Received: from gaia.cc.gatech.edu (reid@gaia.cc.gatech.edu [130.207.3.8]) by burdell.cc.gatech.edu (8.8.Beta.5/8.6.9) with ESMTP id WAA29549; Tue, 5 Nov 1996 22:10:41 -0500 (EST)
	Received: (from reid@localhost) by gaia.cc.gatech.edu (8.8.Beta.5/8.6.9) id WAA00117; Tue, 5 Nov 1996 22:10:40 -0500 (EST)
	From: reid@cc.gatech.edu (Charles Reid Harmon)
	Message-Id: <199611060310.WAA00117@gaia.cc.gatech.edu>
	Subject: Re: Status of SPEC marks, etc.
	To: bill@cc.gatech.edu (Bill Appelbe)
	Date: Tue, 5 Nov 1996 22:10:40 -0500 (EST)
	Cc: annc@cc.gatech.edu, reid@cc.gatech.edu
	In-Reply-To: <199611041604.LAA07415@forge.cc.gatech.edu> from "Bill Appelbe" at Nov 4, 96 11:04:31 am
	X-Mailer: ELM [version 2.4 PL23]
	Content-Type: text
	Status: RO

	> Reid:
	> 	It would be really nice if you could put together a one line
	> description of the status of the SPEC runs on the simulator and compiler, etc.,
	> that we could give to ann's students. Do you have a short summary of
	> how to run the simulator and compiler?

	A small, slightly out-of-date explanation for running the simulator can be
	found as ~reid/superdlx/.documentation.  I'll have to draft something for
	compiling, but they can always examine the Makefile I have for a working
	benchmark: ~reid/superdlx/{CINT95,CFP95}/<benchmark>/Makefile.

	> What i meant was something like:
	> (that would show if they were operational, known problems, feasability
	> of getting them operational (by students with maybe some help from you, etc.)

	Here's the latest on the benchmarks:

	CINT95
	------
	099.go       : working
	124.m88ksim  : not compiled (uses signals)
	126.gcc      : not attempted
	129.compress : working
	130.li       : working
	132.ijpeg    : not working (bizarre runtime problem where the simulator just
	               exits)
	134.perl     : not compiled (requires a slew of unsupported traps)
	147.vortex   : not working (same runtime problem as 132.ijpeg)

	CFP95
	-----
	101.tomcatv  : working
	102.swim     : working
	103.su2cor   : not working (f2c library complains of a formatting error)
	104.hydro2d  : working, but SLOW
	107.mgrid    : working
	110.applu    : working, but SLOW
	125.turb3d   : not working (aborts without error message)
	141.apsi     : not compiled (VAST amount of source code)
	145.fpppp    : not compiled (missing some traps and local fcn cdabs_.  This
	               could likely be fixed easily.)
	146.wave5    : runtime error (DOMAIN error for sqrt)

	Some of these are simulator runtime errors and will definitely require
	simulator patches.  Anything requiring additional traps will also require
	simulator enhancements.  It would be useful to play with some of these
	slow benchmarks and find an input set with a running time in the 1 million
	to 10 million cycle range.  Not a trivial task in most cases.

	Reid



