From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S265375AbTGCUPP (ORCPT ); Thu, 3 Jul 2003 16:15:15 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S265379AbTGCUPP (ORCPT ); Thu, 3 Jul 2003 16:15:15 -0400 Received: from pc2-cwma1-4-cust86.swan.cable.ntl.com ([213.105.254.86]:13465 "EHLO lxorguk.ukuu.org.uk") by vger.kernel.org with ESMTP id S265375AbTGCUPM (ORCPT ); Thu, 3 Jul 2003 16:15:12 -0400 Subject: Re: [RFC] block layer support for DMA IOMMU bypass mode II From: Alan Cox To: Andi Kleen Cc: Grant Grundler , James Bottomley , axboe@suse.de, davem@redhat.com, suparna@in.ibm.com, Linux Kernel Mailing List , alex_williamson@hp.com, bjorn_helgaas@hp.com In-Reply-To: <20030702235619.GA21567@wotan.suse.de> References: <1057077975.2135.54.camel@mulgrave> <20030702015701.6007ac26.ak@suse.de> <20030702165510.GC11739@dsl2.external.hp.com> <1057180598.20318.32.camel@dhcp22.swansea.linux.org.uk> <20030702235619.GA21567@wotan.suse.de> Content-Type: text/plain Content-Transfer-Encoding: 7bit Organization: Message-Id: <1057263988.21508.18.camel@dhcp22.swansea.linux.org.uk> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 (1.2.2-5) Date: 03 Jul 2003 21:26:29 +0100 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Iau, 2003-07-03 at 00:56, Andi Kleen wrote: > > 1. We allocate pages in reverse order so most merges cant occur > > I added an printk and I get quite a lot of merges during bootup > with normal IDE. > > (sometimes 12+ segments) Thats merging adjacent blocks with non adjacent page targets using the IOMMU right - I was doing mergign without an IOMMU which is a little different and turns out to be a waste of cpu