From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id QAA15049; Fri, 13 Aug 2004 16:41:30 +0200 (MET DST) X-Authentication-Warning: pauillac.inria.fr: majordomo set sender to owner-caml-list@pauillac.inria.fr using -f Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id QAA16390 for ; Fri, 13 Aug 2004 16:41:29 +0200 (MET DST) Received: from tkb.mpl.com ([66.109.164.210]) by nez-perce.inria.fr (8.12.10/8.12.10) with ESMTP id i7DEfSmL023603 for ; Fri, 13 Aug 2004 16:41:28 +0200 Received: from tkb.mpl.com (tkb.mpl.com [66.109.164.210]) by tkb.mpl.com (8.12.11/8.12.11) with ESMTP id i7DEfQks079670; Fri, 13 Aug 2004 10:41:27 -0400 (EDT) (envelope-from tkb@tkb.mpl.com) From: "T. Kurt Bond" MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16668.54038.680668.895858@tkb.mpl.com> Date: Fri, 13 Aug 2004 10:41:26 -0400 To: caml-list@inria.fr Cc: "T. Kurt Bond" Subject: Re: AW: [Caml-list] Conservative GC In-Reply-To: <16668.53521.633388.479965@tkb.mpl.com> References: <20040813.125329.74721093.garrigue@kurims.kyoto-u.ac.jp> <411CBAF6.3010101@univ-savoie.fr> <411CC109.6050105@ps.uni-sb.de> <16668.53521.633388.479965@tkb.mpl.com> X-Mailer: VM 7.18 under Emacs 21.3.1 X-Miltered: at nez-perce with ID 411CD318.000 by Joe's j-chkmail (http://j-chkmail.ensmp.fr)! X-Loop: caml-list@inria.fr X-Spam: no; 0.00; caml-list:01 bdw:99 drscheme:01 bdw:99 folks:01 discarded:02 drawback:03 scheme:03 tkb:04 tkb:04 source:07 problem:07 problem:07 difficult:07 occasionally:07 Sender: owner-caml-list@pauillac.inria.fr Precedence: bulk Conservative GC has the drawback that there *are* pathological cases where the GC, because of its conservatism, retains memory that could actually be discarded. If I remember correctly, the PLT Scheme folks, who use the BDW conservative gc, apparently ran into this with the web server that is distributed with DrScheme, and I've seen mention of other programs that use the BDW gc occasionally running into this problem. In these cases it is often possible to tune the program to eliminate the pathological case, but it's often difficult to figure out the source of the problem. Accurate (that is, non-conservative) gc, of course, does not have this problem. -- T. Kurt Bond, tkb@tkb.mpl.com ------------------- To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/ Beginner's list: http://groups.yahoo.com/group/ocaml_beginners