Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Mar 2004 12:55:10 -0700 (MST)
From:      RJ45 <rj45@slacknet.com>
To:        freebsd-alpha@freebsd.org
Subject:   ccc optimizations ?
Message-ID:  <Pine.LNX.4.21.0403121247430.3880-100000@slacknet.slacknet.com>

next in thread | raw e-mail | index | archive | help



Hello,
what are the best comand line options to gain a fastest as possible
executable using ccc ?

I have a XP1000 with 4.9 release ( with 5.x I had lots a problems with
ccc).

I tryed these options:

ccc -fast -O3 -arch ev67 -ifo

and I gaing slightly better results than using ccc with no options

actually after many tries it looks like "-fast" is the one option
which counts something, all the others (-O3 -arch or -tune)
looks like useless when used with -fast

if I don't use -fast also I have to compile the ieee_fp_control.c and
errno_location.c together with my code otherwise ther eare unresolved
externals.

anyone has some feedback about how to gain the best performances
out of ccc ?

Anyway I want to mention that with a progam doing lots of integer
calculations and iterations with arctan function on my XP1000
using ccc I get better results that on a Pentium IV 3GHz using gcc
with i686 optimization switches.
ev6 is a great CPU.

Rick




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0403121247430.3880-100000>