From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.1.3 (2006-06-01) on yquem.inria.fr X-Spam-Level: X-Spam-Status: No, score=0.0 required=5.0 tests=AWL autolearn=disabled version=3.1.3 X-Original-To: caml-list@yquem.inria.fr Delivered-To: caml-list@yquem.inria.fr Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by yquem.inria.fr (Postfix) with ESMTP id 409FEBC69 for ; Mon, 25 Sep 2006 15:40:14 +0200 (CEST) Received: from smtp3.adl2.internode.on.net (smtp3.adl2.internode.on.net [203.16.214.203]) by concorde.inria.fr (8.13.6/8.13.6) with ESMTP id k8PDeB7m028513 for ; Mon, 25 Sep 2006 15:40:13 +0200 Received: from rosella (ppp14-47.lns2.syd7.internode.on.net [59.167.14.47]) by smtp3.adl2.internode.on.net (8.13.6/8.13.5) with ESMTP id k8PDe2fY006844; Mon, 25 Sep 2006 23:10:03 +0930 (CST) (envelope-from skaller@users.sourceforge.net) Subject: Re: [Caml-list] Regarding SMP computing From: skaller To: Jacques Carette Cc: caml-list@inria.fr In-Reply-To: <4517C7FE.8080102@mcmaster.ca> References: <4517C7FE.8080102@mcmaster.ca> Content-Type: text/plain Date: Mon, 25 Sep 2006 23:40:02 +1000 Message-Id: <1159191602.5339.98.camel@rosella.wigram> Mime-Version: 1.0 X-Mailer: Evolution 2.6.1 Content-Transfer-Encoding: 7bit X-Miltered: at concorde with ID 4517DC3B.000 by Joe's j-chkmail (http://j-chkmail.ensmp.fr)! X-Spam: no; 0.00; peyton-jones:01 heap:01 compaction:01 heap:01 threads:01 2006:98 sourceforge:01 wrote:01 garbage:01 minor:01 minor:01 caml-list:01 ghc:01 ghc:01 jacques:03 On Mon, 2006-09-25 at 08:13 -0400, Jacques Carette wrote: > Over on Haskell-cafe, Simon Peyton-Jones says: > "GHC 6.6 (release candidate available) supports parallel execution on > SMP machines. > > Garbage collection is not parallelised yet, something we plan to fix > this autumn." > > A bit of competition is a good thing, isn't it? Actually this isn't quite accurate, at least as I understand it. GHC already does have fully parallel minor heap allocation and compaction etc using a per-thread minor heap, but the major collection isn't only not parallel .. it isn't even concurrent. [Using Cheng's terminology, parallel collection implies multiple collector threads can cooperate so that the system can scale to very large numbers of CPUs by allowing more than one CPU to run the collector in simultaneously, 'merely' concurrent includes collectors limited to a single collector thread] -- John Skaller Felix, successor to C++: http://felix.sf.net