From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754847AbYIWCvT (ORCPT ); Mon, 22 Sep 2008 22:51:19 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754344AbYIWCvJ (ORCPT ); Mon, 22 Sep 2008 22:51:09 -0400 Received: from casper.infradead.org ([85.118.1.10]:55695 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754175AbYIWCvI (ORCPT ); Mon, 22 Sep 2008 22:51:08 -0400 Date: Mon, 22 Sep 2008 19:50:59 -0700 From: Arjan van de Ven To: Jeremy Fitzhardinge Cc: Joerg Roedel , linux-kernel@vger.kernel.org, kvm@vger.kernel.org, iommu@lists.linux-foundation.org, David Woodhouse , Muli Ben-Yehuda , Amit Shah , Ingo Molnar , FUJITA Tomonori Subject: Re: [PATCH 0/9][RFC] stackable dma_ops for x86 Message-ID: <20080922195059.008de827@infradead.org> In-Reply-To: <48D85758.7080303@goop.org> References: <1222107681-8185-1-git-send-email-joerg.roedel@amd.com> <20080922113619.5075e7e4@infradead.org> <20080922183944.GJ24392@amd.com> <48D85758.7080303@goop.org> Organization: Intel X-Mailer: Claws Mail 3.5.0 (GTK+ 2.12.11; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-SRS-Rewrite: SMTP reverse-path rewritten from by casper.infradead.org See http://www.infradead.org/rpr.html Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 22 Sep 2008 19:41:28 -0700 Jeremy Fitzhardinge wrote: > Joerg Roedel wrote: > > Its implemented using the per-device dma-ops already there. With > > this patches there is a list of available dma_ops implementations > > which are asked in a particular order if they can handle the > > device. The first implementation which returns true is assigned to > > the device as the per-device dma_ops structure. > > > > (Hmm, maybe the name stackable is misleading, is "dma_ops > > multiplexing" better?) > > Is per-device the right level? Wouldn't per-bus make more sense? not really; all DMA functions get a device as argument already anyway; just going to bus makes no sense there. Even if you set it the same for the whole bus almost all of the time... the APIs just work per device. (and device assignment clearly is per device as well) -- Arjan van de Ven Intel Open Source Technology Centre For development, discussion and tips for power savings, visit http://www.lesswatts.org