Path: utzoo!utgpu!cunews!bnrgate!brtph3!brchh104!brchs1!bnr.ca!rice.edu!sun-spots-request
From: slevy@poincare.geom.umn.edu (Stuart Levy)
Newsgroups: comp.sys.sun
Subject: Re: GNU cc vs cc
Keywords: Software
Message-ID: <487@brchh104.bnr.ca>
Date: 28 Nov 90 17:45:45 GMT
Sender: news@brchh104.bnr.ca
Organization: Sun-Spots
Lines: 14
Approved: Sun-Spots@rice.edu
X-Original-Date: Wed, 21 Nov 90 02:53:03 GMT
X-Refs:  Original: v9n366
X-Sun-Spots-Digest: Volume 9, Issue 375, message 6
X-Note: Submissions: sun-spots@rice.edu, Admin: sun-spots-request@rice.edu

In article <363@brchh104.bnr.ca> HORN%HYDRA@sdi.polaroid.com writes:
>I recently conducted a more realistic test of the compilers and found gcc
>version 1.37.1 significantly faster than Sun cc from 4.0.3 on a SS1.  The
>test was TeX compiled gcc -O -finline... -fstrength... versus cc -O4. ...
>  Drhystones are often misleading.

You have to be careful about what is realistic though... Trying some
homebrew DES-encryption code with cc -O4 vs. gcc 1.37.1, on a Sparcstation
under both 4.0.3 and 4.1, Sun cc won each time by about 10% on both of two
DES-related operations.  TeX is certainly much bigger, but both are "real"
applications.

    Stuart Levy, Geometry Group, University of Minnesota
    slevy@geom.umn.edu