Xref: utzoo comp.arch:19957 comp.sys.sgi:7500
Path: utzoo!attcan!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!agate!pasteur!sprite.berkeley.edu!tve
From: tve@sprite.berkeley.edu (Thorsten von Eicken)
Newsgroups: comp.arch,comp.sys.sgi
Subject: Re: reliable/reproduceable benchmarks on SGI MIPS box
Message-ID: <9932@pasteur.Berkeley.EDU>
Date: 24 Dec 90 08:43:58 GMT
References: <11737@alice.att.com>
Sender: news@pasteur.Berkeley.EDU
Reply-To: tve@sprite.berkeley.edu (Thorsten von Eicken)
Organization: University of California, Berkeley
Lines: 8

... quick 2 cents worth of guesses:
you haven't said whether you're running your program on all 8 processors
or on only one of them. if you're running on only one, could it be that
the other seven interfere? What happens if you run a "for(;;);" program
on seven processors while running the benchmark on the eighth?
also, is there a cache-flush system call you can call before starting the
timer?
	TvE