*BSD News Article 73889


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mira.net.au!news.mel.connect.com.au!munnari.OZ.AU!spool.mu.edu!howland.reston.ans.net!newsfeed.internetmci.com!zdc!szdc!szdc-e!news
From: "John S. Dyson" <toor@dyson.iquest.net>
Newsgroups: comp.os.linux.networking,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc
Subject: Re: TCP latency
Date: Tue, 16 Jul 1996 10:07:03 -0500
Organization: John S. Dyson's home machine
Lines: 68
Message-ID: <31EBB017.167EB0E7@dyson.iquest.net>
References: <4paedl$4bm@engnews2.Eng.Sun.COM> <31E7C0DD.41C67EA6@dyson.iquest.net> <4s8tcn$jsh@fido.asd.sgi.com> <31E80ACA.167EB0E7@dyson.iquest.net> <4sadde$qsv@linux.cs.Helsinki.FI> <31E9E3A7.41C67EA6@dyson.iquest.net> <4sefde$f0l@fido.asd.sgi.com>
NNTP-Posting-Host: dyson.iquest.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.0b5aGold (X11; I; FreeBSD 2.2-CURRENT i386)
Xref: euryale.cc.adfa.oz.au comp.os.linux.networking:45437 comp.unix.bsd.netbsd.misc:4077 comp.unix.bsd.freebsd.misc:23753

Larry McVoy wrote:
> 
> John S. Dyson (toor@dyson.iquest.net) wrote:
> : We could have special cased it to make the benchmark look better, but
> : we did not.  We have looked at it carefully, and it has little impact
> : on real-world performance.  Historically, a NULL syscall is not
> : a write to /dev/null...  That is what the benchmark is, and simply
> : it is NOT IMPORTANT.
> 
> Umm, I'd be happy to entertain suggestions for a better measurement of
> a null entry into the system.  I don't want something that anyone special
> cases - that's just worthless.  I want something that is actually measuring
> all the work you need to do to get to the point that you can do something in
> the kernel.
>
> 
> Maybe gettimeofday() would have been a better null syscall, what do you think?
> 
 
Perhaps clearer naming.  Hows about gettimeofday() for a null syscall,
and
specifically measure VFS, socket, whatever overhead.  I really don't
have
a problem with the benchmark that you call "lat_syscall", but it is
different
that I would expect.  My initial response was "what?", but I do
understand
your reasoning (I think.)  The VFS on FreeBSD had the layering added to
it,
and I think that is where the unexpected overhead is.  Unfortunately, it
is still up-in-the-air what we are going to do with it, and so I don't
want
to make it faster until we know (and then become more incompatible with
the
other *BSDs.)  So, it does appear that 'lat_syscall'
inadvertently tickled a soft spot in FreeBSD right now.  I do believe
that
the FreeBSD syscall overhead is greater than Linux, but not by the
difference measured by the current 'lat_syscall'.

> 
> Well, I doubt you'll buy it, but I agree with Linus that your message
> certainly implied that Linux special cased the null syscal metric used
> in lmbench.  IF that was your intended implication, you may not have been
> lying, but you certainly were implying something that simply isn't true.
>
Firstly, you are one of the first people that I have seen who apparently
understands that a lie requires intent.  Additionally, I was speaking of
something inside of ME.  I did not do it.  So I did not lie because
there
was absolutely no intent, there was no implication, and there was an
inferrence that had nothing to do with the facts.  (Note the difference
between imply (that is what I do) and infer (that is what the listener
does)).
> 
> : Linus, you are an arrogant, self-righteous a**hole...
> 
> Umm, when it gets to this level, maybe it is better to give yourself a
> day or two before you hit that send button.  Do you really want to
> make public statements like this?  Seems a little extreme to me.
>
LIAR is the worst thing that I can be called.  I think that Linus
needs to be careful about calling people liars.  How does he
know that there was intent?  It requires serious amounts of
ego for him to be able to see what 'intent' that I have.

John