From owner-freebsd-arch@FreeBSD.ORG Tue Apr 25 02:39:30 2006 Return-Path: X-Original-To: arch@freebsd.org Delivered-To: freebsd-arch@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 70BD016A400 for ; Tue, 25 Apr 2006 02:39:30 +0000 (UTC) (envelope-from gnn@neville-neil.com) Received: from mrout2-b.corp.dcn.yahoo.com (mrout2-b.corp.dcn.yahoo.com [216.109.112.28]) by mx1.FreeBSD.org (Postfix) with ESMTP id 171D443D46 for ; Tue, 25 Apr 2006 02:39:30 +0000 (GMT) (envelope-from gnn@neville-neil.com) Received: from minion.local.neville-neil.com (proxy7.corp.yahoo.com [216.145.48.98]) by mrout2-b.corp.dcn.yahoo.com (8.13.6/8.13.4/y.out) with ESMTP id k3P2cuiK039308; Mon, 24 Apr 2006 19:38:56 -0700 (PDT) Date: Tue, 25 Apr 2006 10:22:54 +0900 Message-ID: From: gnn@freebsd.org To: Kirk McKusick In-Reply-To: <200604240633.k3O6XUJ0042841@chez.mckusick.com> References: <200604240633.k3O6XUJ0042841@chez.mckusick.com> User-Agent: Wanderlust/2.14.0 (Africa) SEMI/1.14.6 (Maruoka) FLIM/1.14.8 (=?ISO-8859-4?Q?Shij=F2?=) APEL/10.6 Emacs/22.0.50 (i386-apple-darwin8.5.1) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Cc: arch@freebsd.org Subject: Re: Linus Torvalds on FreeBSD's Use of Copy-on-write X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Apr 2006 02:39:30 -0000 At Sun, 23 Apr 2006 23:33:30 -0700, Kirk McKusick wrote: > > Anyone working on zero-copy sockets care to respond to this? > > http://developers.slashdot.org/developers/06/04/21/1536213.shtml > > Linus Torvalds made reference to some possible future extensions. > This included vmsplice(), a system call since implemented by Jens > Axboe "to basically do a 'write to the buffer', but using the > reference counting and VM traversal to actually fill the buffer." > Reviewing the implications of using such a system call lead to a > comparison with FreeBSD's ZERO_COPY_SOCKET which uses COW (copy on > write). > To be honest, what all this has made me think of is that we should have a "big board" of unsolved problems we'd like to look at. This certainly deserves a place, as it's an interesting question. As scientists and engineers we should be interested in such things, even if the original statement was poorly worded. Perhaps on a FreeBSD Feature Wiki? We could even propose a benchmark as a SoC project. Later, George