From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754036AbZLBIIF (ORCPT ); Wed, 2 Dec 2009 03:08:05 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751793AbZLBIIE (ORCPT ); Wed, 2 Dec 2009 03:08:04 -0500 Received: from sh.osrg.net ([192.16.179.4]:50617 "EHLO sh.osrg.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751791AbZLBIID (ORCPT ); Wed, 2 Dec 2009 03:08:03 -0500 Date: Wed, 2 Dec 2009 17:07:24 +0900 To: yinghai@kernel.org Cc: fujita.tomonori@lab.ntt.co.jp, mingo@elte.hu, linux-kernel@vger.kernel.org Subject: Re: [PATCH -tip] x86: fix iommu=soft boot option From: FUJITA Tomonori In-Reply-To: <4B161D70.3030301@kernel.org> References: <4B160FE1.1090307@kernel.org> <20091202162510Q.fujita.tomonori@lab.ntt.co.jp> <4B161D70.3030301@kernel.org> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: <20091202170715K.fujita.tomonori@lab.ntt.co.jp> X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-3.0 (sh.osrg.net [192.16.179.4]); Wed, 02 Dec 2009 17:07:24 +0900 (JST) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 01 Dec 2009 23:55:28 -0800 Yinghai Lu wrote: > >>>> amd 64 systems that > >>>> 1. do not have AGP > >>>> 2. do not have IOMMU > >>>> 3. mem > 4g > >>>> 4. BIOS do not allocate correct gart in NB. > >>>> will leave them to use SWIOTLB forcely. > >>> As I asked earlier, can you tell me what dma ops such system is > >>> supposed to use? > >> gart_dma_ops > > > > How does gart_dma_ops work on systems without IOMMU? > > ? If a system doesn't have IOMMU, what does gart_dma_ops do? > >>>> -v4: disable translating in early_gart_iommu_check according to FUJITA > >>> Why can't we always disable translation in early_gart_iommu_check? We > >>> really need search_agp_bridge() in early_gart_iommu_check()? > >> just don't want to messed them up. > > > > Sorry, I can't follow you. What does 'always disabling' mess up? > > the AMD K8 system with AGP etc...those systems are some kind of 5 years old. > > and don't have those kind of system to verify... Ok.