From owner-freebsd-hackers Sun Feb 3 9:29:18 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from math.smsu.edu (math.smsu.edu [146.7.45.198]) by hub.freebsd.org (Postfix) with ESMTP id 1C7D237B400 for ; Sun, 3 Feb 2002 09:29:16 -0800 (PST) Received: by math.smsu.edu (Postfix, from userid 1000) id 4CFFB8E118; Sun, 3 Feb 2002 11:29:16 -0600 (CST) To: freebsd-hackers@freebsd.org Subject: Re: unbelievable benchmark output Message-Id: <20020203172916.4CFFB8E118@math.smsu.edu> Date: Sun, 3 Feb 2002 11:29:16 -0600 (CST) From: erik@math.smsu.edu (Erik Greenwald) Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG > > Several runs of the program take about the same time but the time > changes wildly when the executable is called differently. > > ------- > ./xx/xxx > 5 s > xx/xxx > 9 s > > The only thing which I can think of that can be causing this is some > memory alignment issue. > it could also be the shell taking time to look up the names? > Compile the code below with gcc -O -march=pentiumpro and save the > program to /tmp/x (1x) and /tmp/xxxxxxxxxxx (11x). cd /tmp; ./x shows me > 0.490495 (with slight variations amongst runs) and ./xxxxxxxxxxx 1.016591. > The real results depend on hw but they occur always with different > differences in speed and the length of name of fast/slow programs. > here're my numbers athlon 850, iwill k266 (via686b southbridge), 384m ram, 5400rpm fujitsu 20G FreeBSD fenris 5.0-CURRENT FreeBSD 5.0-CURRENT #10: Fri Feb 1 23:57:28 CST 2002 root@fenris:/usr/src/sys/i386/compile/FENRIS i386 ./x bounces between .6 and .7 ./xxx...xx bounces between .65 and .75 with 100 x's in the name, .67 to .76 /tmp/x .55 .60 /tmp/x(11) .62 /tmp/x(100) .62 .72 > > > -- > Michal Mertl > mime@traveller.cz > -Erik [http://math.smsu.edu/~erik] The opinions expressed by me are not necessarily opinions. In all probability, they are random rambling, and to be ignored. Failure to ignore may result in severe boredom or confusion. Shake well before opening. Keep Refrigerated. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message