From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: [PATCH] dma-debug: disable DMA_API_DEBUG for now Date: Fri, 5 Jun 2009 15:57:56 +0100 Message-ID: <200906051557.58809.arnd@arndb.de> References: <20090605173232N.fujita.tomonori@lab.ntt.co.jp> <20090605104132.GE24836@amd.com> <20090605203753K.fujita.tomonori@lab.ntt.co.jp> Mime-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from moutng.kundenserver.de ([212.227.126.186]:58736 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751798AbZFEO60 (ORCPT ); Fri, 5 Jun 2009 10:58:26 -0400 In-Reply-To: <20090605203753K.fujita.tomonori@lab.ntt.co.jp> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: FUJITA Tomonori Cc: joerg.roedel@amd.com, torvalds@linux-foundation.org, mingo@elte.hu, lethal@linux-sh.org, just.for.lkml@googlemail.com, hancockrwd@gmail.com, jens.axboe@oracle.com, bharrosh@panasas.com, linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org On Friday 05 June 2009, FUJITA Tomonori wrote: > I think that the better fix can handle both cases per device: > > - multiple identical dma addresses should not happen (with devices > behind hardware IOMMU) > - multiple identical dma addresses could happen I guess you could also have the case where for a given range of addresses, you use a linear mapping and the dma addresses can be identical, while for other physical addresses you would rely on address translation. For example on PowerPC/Cell with infiniband adapters, you can get linear mapping behavior for DMA_ATTR_WEAK_ORDERING but IOMMU translation without that flag, for the same device and same physical address. Arnd <><