*BSD News Article 78439


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.uwa.edu.au!disco.iinet.net.au!wa.news.telstra.net!act.news.telstra.net!psgrain!usenet.eel.ufl.edu!news.mathworks.com!enews.sgi.com!news.sgi.com!news-out.microserve.net!news-in.microserve.net!zdc-e!super.zippo.com!zdc!szdc!szdc-e!news
From: John Dyson <dyson@freebsd.org>
Newsgroups: comp.unix.bsd.freebsd.misc,comp.os.linux.misc
Subject: Re: Benchmarking different Unix Operating Systems
Date: Sun, 15 Sep 1996 14:19:50 -0500
Organization: John S. Dyson's home machine
Lines: 25
Message-ID: <323C56D6.41C67EA6@freebsd.org>
References: <aak2.842008017@Isis.MsState.Edu> <50p41e$1ie@nnrp1.news.primenet.com> <32305b37.590852758@news.intellistar.net> <aak2.842255270@Isis.MsState.Edu> <511pa6$avb@agate.berkeley.edu> <513f4t$ac@anorak.coverform.lan> <323652FA.41C67EA6@freebsd.org> <51gmm2$snk@josie.abo.fi>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.0Gold (X11; U; FreeBSD 2.2-CURRENT i386)
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:27401 comp.os.linux.misc:129663

Mats Andtbacka wrote:
> 
> John Dyson, in <323652FA.41C67EA6@freebsd.org>:
> [...]
> >Also, the FreeBSD that they benchmarked is the "old stuff".  Even though
> >it isn't suggested -- there are people using FreeBSD-current in
> >production, and certain vintages of it are very good (and light
> >years beyond 2.1.X.)  And as always, FreeBSD is never compiled
> >-fomit-frame-pointer or with optimizations over -O out of the box.
> 
> out of interest, why is this thus? i assume it must be a deliberate
> policy decision, but what might be its background?
>
-fomit-frame-pointer messes up stack tracebacks.
Greater than or equal to -O2 would be nice, but the compiler reliability
becomes progressively more questionable. Sad thing is that
-fomit-frame-pointer seems to make alot of difference, but the policy
of being able to do full tracebacks overrules this.  You can
recompile your kernel with better optimization.  It does disadvantage
FreeBSD in benchmarks, but we would rather make the kernel more
useful.  In my eyes, quality is not just low level/unloaded speed.


John