From owner-freebsd-hackers  Sun Apr 26 03:25:08 1998
Return-Path: <owner-freebsd-hackers@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id DAA11720
          for freebsd-hackers-outgoing; Sun, 26 Apr 1998 03:07:06 -0700 (PDT)
          (envelope-from owner-freebsd-hackers@FreeBSD.ORG)
Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA10423
          for <freebsd-hackers@freebsd.org>; Sun, 26 Apr 1998 02:57:15 -0700 (PDT)
          (envelope-from julian@whistle.com)
Received: (from daemon@localhost)
	by alpo.whistle.com (8.8.5/8.8.5) id CAA06772;
	Sun, 26 Apr 1998 02:53:14 -0700 (PDT)
Received: from current1.whistle.com(207.76.205.22)
 via SMTP by alpo.whistle.com, id smtpd006768; Sun Apr 26 09:53:09 1998
Date: Sun, 26 Apr 1998 02:47:44 -0700 (PDT)
From: Julian Elischer <julian@whistle.com>
To: Tom <tom@sdf.com>
cc: Christoph Toshok <toshok@Hungry.COM>, freebsd-hackers@FreeBSD.ORG
Subject: Re: threads performance
In-Reply-To: <Pine.BSF.3.95q.980426012031.15432B-100000@misery.sdf.com>
Message-ID: <Pine.BSF.3.95.980426024647.9124B-100000@current1.whistle.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

try KTRACEing the two versions.
I'll bet you'll see the difference immediatly.


On Sun, 26 Apr 1998, Tom wrote:

> 
> On 25 Apr 1998, Christoph Toshok wrote:
> 
> > Are there any plans to address the performance of threads in the
> > coming weeks/months?  The fact that NSPR can drop 21 seconds off the
> > runtime (in this very contrived example) makes me think that there is
> > a lot going on in libc_r that is suboptimal, but perhaps there is just
> > no other way to implement things so they conform to the posix spec.
> 
>   Even mit-pthreads on a 2.2.6 system is faster than libc_r (using the
> tests in mysql as a comparison).
> 
>   The weird part is that the amount of CPU time accumulated is very
> similar, libc_r just takes more real time.  Makes me think that something
> in libc_r just sleeps once and while...
> 
> Tom
> 
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-hackers" in the body of the message
> 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message