From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751233AbWDKFtt (ORCPT ); Tue, 11 Apr 2006 01:49:49 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932099AbWDKFtt (ORCPT ); Tue, 11 Apr 2006 01:49:49 -0400 Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:45491 "EHLO sunset.davemloft.net") by vger.kernel.org with ESMTP id S1751233AbWDKFts (ORCPT ); Tue, 11 Apr 2006 01:49:48 -0400 Date: Mon, 10 Apr 2006 22:49:33 -0700 (PDT) Message-Id: <20060410.224933.39567033.davem@davemloft.net> To: benh@kernel.crashing.org Cc: benoit.boissinot@ens-lyon.org, mb@bu3sch.de, netdev@vger.kernel.org, bcm43xx-dev@lists.berlios.de, linux-kernel@vger.kernel.org, linville@tuxdriver.com Subject: Re: [RFC/PATCH] remove unneeded check in bcm43xx From: "David S. Miller" In-Reply-To: <1144719972.19353.24.camel@localhost.localdomain> References: <200604100607.33362.mb@bu3sch.de> <20060410042228.GN27596@ens-lyon.fr> <1144719972.19353.24.camel@localhost.localdomain> X-Mailer: Mew version 4.2.53 on Emacs 21.4 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Benjamin Herrenschmidt Date: Tue, 11 Apr 2006 11:46:12 +1000 > But ppc64 hits the problem and at this point, there is nothing > I can do other than either implementing a split zone allocation mecanism > in the ppc64 architecture for the sole sake of bcm43xx (ick !) or doing > some trick with the iommu... I think allowing DMA mask range limiting in the IOMMU layer is going to set a very bad precedence, just don't do it. It's 2006, we should be way past the era of not putting the full 32 PCI DMA address bits in devices. In this day and age it is simply inexscusable. Maybe we could understand chips coming out 8 years ago when a lot of designs were transitioning from ISA to PCI, but that no longer applies in any way today.