From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dsl027-180-168.sfo1.dsl.speakeasy.net ([216.27.180.168]:10664 "EHLO sunset.davemloft.net") by vger.kernel.org with ESMTP id S1751485AbWEUFz0 (ORCPT ); Sun, 21 May 2006 01:55:26 -0400 Date: Sat, 20 May 2006 22:55:30 -0700 (PDT) Message-Id: <20060520.225530.23952587.davem@davemloft.net> Subject: Re: Problem with sg_dma_address() on ia64, sh and sh64 From: "David S. Miller" In-Reply-To: <1148141933.3529.25.camel@mulgrave.il.steeleye.com> References: <1148141933.3529.25.camel@mulgrave.il.steeleye.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org To: James.Bottomley@steeleye.com Cc: linux-arch@vger.kernel.org List-ID: From: James Bottomley Date: Sat, 20 May 2006 11:18:53 -0500 > This is the defined accessor to our struct scatterlist. These three > architectures define it in asm/pci.h rather than asm/scatterlist.h. > This means that one doesn't get the accessor simply by including > linux/scatterlist.h, which one should. This means that non-pci uses > don't work (like our budding libsrp). > > Could you all move it to where it should be? I agree, there should be nothing PCI specific about the scatterlist DMA entry accessors.