From owner-freebsd-hackers@FreeBSD.ORG Wed Jun 2 15:29:47 2004 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A5A3816A4CE; Wed, 2 Jun 2004 15:29:47 -0700 (PDT) Received: from xorpc.icir.org (xorpc.icir.org [192.150.187.68]) by mx1.FreeBSD.org (Postfix) with ESMTP id 90A8E43D2F; Wed, 2 Jun 2004 15:29:47 -0700 (PDT) (envelope-from rizzo@icir.org) Received: from xorpc.icir.org (localhost [127.0.0.1]) by xorpc.icir.org (8.12.9p1/8.12.8) with ESMTP id i52MTlgd017656; Wed, 2 Jun 2004 15:29:47 -0700 (PDT) (envelope-from rizzo@xorpc.icir.org) Received: (from rizzo@localhost) by xorpc.icir.org (8.12.9p1/8.12.3/Submit) id i52MTl9i017655; Wed, 2 Jun 2004 15:29:47 -0700 (PDT) (envelope-from rizzo) Date: Wed, 2 Jun 2004 15:29:47 -0700 From: Luigi Rizzo To: "Christian S.J. Peron" Message-ID: <20040602152947.B17332@xorpc.icir.org> References: <20040602043537.GA42327@freefall.freebsd.org> <20040602135155.GA31642@freefall.freebsd.org> <20040602213515.GA90619@freefall.freebsd.org> <40BE4BFE.70204@freebsd.org> <20040602221443.GA92431@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <20040602221443.GA92431@freefall.freebsd.org>; from csjp@freebsd.org on Wed, Jun 02, 2004 at 03:14:43PM -0700 cc: hackers@freebsd.org cc: Andre Oppermann cc: ipfw@freebsd.org Subject: Re: ipfw cached ucred patch 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, 02 Jun 2004 22:29:47 -0000 On Wed, Jun 02, 2004 at 03:14:43PM -0700, Christian S.J. Peron wrote: > > I understand what you are saying. The only real other choice > would be to copy out the entire cr_groups array. Do you know > if this copy would be more expensive then the mutex lock/unlock > associated with grabbing a reference to the ucred? i bet the copy it would be cheaper almost on any architecture -- it is only 64 bytes anyways, with these sizes what kills you in memory accesses is the latency, not the throughput. cheers luigi