From owner-freebsd-hackers@FreeBSD.ORG Wed Feb 16 04:11:58 2005 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from green.homeunix.org (freefall.freebsd.org [216.136.204.21]) by hub.freebsd.org (Postfix) with ESMTP id BBB8E16A4CE; Wed, 16 Feb 2005 04:11:57 +0000 (GMT) Received: from green.homeunix.org (green@localhost [127.0.0.1]) by green.homeunix.org (8.13.1/8.13.1) with ESMTP id j1G4BtaJ033965; Tue, 15 Feb 2005 23:11:55 -0500 (EST) (envelope-from green@green.homeunix.org) Received: (from green@localhost) by green.homeunix.org (8.13.1/8.13.1/Submit) id j1G4BrPJ033964; Tue, 15 Feb 2005 23:11:53 -0500 (EST) (envelope-from green) Date: Tue, 15 Feb 2005 23:11:53 -0500 From: Brian Fundakowski Feldman To: Andrew MacIntyre , Jason Henson , freebsd-hackers@FreeBSD.ORG Message-ID: <20050216041153.GC1069@green.homeunix.org> References: <1108277558l.86500l.0l@BARTON> <20050213082128.GA68307@VARK.MIT.EDU> <42108243.9030800@bullseye.apana.org.au> <20050214130450.GA55300@VARK.MIT.EDU> <20050214220807.GF40468@funkthat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050214220807.GF40468@funkthat.com> User-Agent: Mutt/1.5.6i Subject: Re: malloc vs ptmalloc2 X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Feb 2005 04:11:58 -0000 On Mon, Feb 14, 2005 at 02:08:07PM -0800, John-Mark Gurney wrote: > David Schultz wrote this message on Mon, Feb 14, 2005 at 08:04 -0500: > > Right, databases, language runtimes, and the small set of other > > applications for which it really matters usually have their own > > special-purpose allocators. I was counting on that when I said > > that replacing malloc() is unlikely to make a big difference. > > (One could argue, of course, that it's unfortunate that > > applications need to do so.) > > Hmmm, maybe we should bring the zone API into userland? :) It'd be > nice to have some of the features, like the ctor/dtor's... Why not just make UMA the userland implementation, too? It should only cost us one kernel trap instruction per allocation/deallocation operation. With fine-grained kernel locking, why we wouldn't want to trust the same code to both tasks, and possibly optimize more because it can do that? -- Brian Fundakowski Feldman \'[ FreeBSD ]''''''''''\ <> green@FreeBSD.org \ The Power to Serve! \ Opinions expressed are my own. \,,,,,,,,,,,,,,,,,,,,,,\