From xemacs-m  Mon Sep 22 18:39:28 1997
Received: from relay3.UU.NET (relay3.UU.NET [192.48.96.8])
	by xemacs.org (8.8.5/8.8.5) with ESMTP id SAA11424
	for <xemacs-beta@xemacs.org>; Mon, 22 Sep 1997 18:39:27 -0500 (CDT)
Received: from crystal.WonderWorks.COM by relay3.UU.NET with ESMTP 
	(peer crosschecked as: crystal.WonderWorks.com [192.203.206.1])
	id QQdidy01905; Mon, 22 Sep 1997 19:39:37 -0400 (EDT)
Received: by crystal.WonderWorks.COM 
	id QQdidy13097; Mon, 22 Sep 1997 19:39:17 -0400 (EDT)
Date: Mon, 22 Sep 1997 19:39:17 -0400 (EDT)
Message-Id: <QQdidy13097.199709222339@crystal.WonderWorks.COM>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Kyle Jones <kyle_jones@wonderworks.com>
To: XEmacs Developers <xemacs-beta@xemacs.org>
Subject: Re: dlmalloc
In-Reply-To: <kigzpp5dvzj.fsf@jagor.srce.hr>
References: <m27mcame88.fsf@altair.xemacs.org>
	<kig7mc91r56.fsf@jagor.srce.hr>
	<87wwk9yz85.fsf_-_@bittersweet.inetarena.com>
	<kigzpp5dvzj.fsf@jagor.srce.hr>
X-Mailer: VM 6.34 under Emacs 19.34.2
X-Face: /cA45WHG7jWq>(O3&Z57Y<"WsX5ddc,4c#w0F*zrV#=M
        0@~@,s;b,aMtR5Sqs"+nU.z^CSFQ9t`z2>W,S,]:[+2^
        Nbf6v4g>!&,7R4Ot4Wg{&tm=WX7P["9%a)_da48-^tGy
        ,qz]Z,Zz\{E.,]'EO+F)@$KtF&V

Hrvoje Niksic writes:
 > karlheg@inetarena.com (Karl M. Hegbloom) writes:
 > 
 > >     Hrvoje> Should we add the same stuff to XEmacs?  What *is*
 > >     Hrvoje> DOUG_LEA_MALLOC anyway?  We need a malloc guru here.
 > > 
 > >  I'm not a malloc guru... but I know that the Doug Lea malloc is
 > >  what's in the Linux glibc-2.  XEmacs won't work if you tell configure
 > >  to use-system-malloc.  I wonder if that patch would fix it?
 > 
 > Maybe.  Either way, that wasn't what I hoped it would be (an
 > optimization for consing to explain the difference between 19.34 and
 > 20.1)

Steven mentioned that 19.34 was linked with a different version
of the C library than the v20 Emacses he tested.  That makes me
distrust the 19.34 numbers as a comparative reference if you're
looking for faster ways to do a particular thing.  On my machine,
using the same library and compiler flags for all builds, 19.34
benches only 6% faster than XEmacs-20.3-Latin1 on the small lists
test.  Emacs 20.2 benches much slower than both of them.  I ran
the test a few times and took the best result for each Emacs
type.

Emacs 19.34	5.80
XEmacs 20.3	6.10
Emacs 20.1	7.76

