From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S266626AbUHIO07 (ORCPT ); Mon, 9 Aug 2004 10:26:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S266613AbUHIOZW (ORCPT ); Mon, 9 Aug 2004 10:25:22 -0400 Received: from ns.virtualhost.dk ([195.184.98.160]:51917 "EHLO virtualhost.dk") by vger.kernel.org with ESMTP id S266626AbUHIOWG (ORCPT ); Mon, 9 Aug 2004 10:22:06 -0400 Date: Mon, 9 Aug 2004 16:21:38 +0200 From: Jens Axboe To: Joerg Schilling Cc: paul@clubi.ie, James.Bottomley@steeleye.com, linux-kernel@vger.kernel.org, mj@ucw.cz Subject: Re: PATCH: cdrecord: avoiding scsi device numbering for ide devices Message-ID: <20040809142137.GR10418@suse.de> References: <200408091413.i79EDt9L010562@burner.fokus.fraunhofer.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200408091413.i79EDt9L010562@burner.fokus.fraunhofer.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Aug 09 2004, Joerg Schilling wrote: > > >From: Paul Jakma > > >> Of course, ATAPI devices on Solaris are handled by the same > >> target drivers as e.g. those on 50 pin cables. > > >Yes ATAPI is. > > >> The ATA driver is implemented the way one would expect it by acting > >> as a SCSI HBA. > > >Yes, as does libata on Linux. > > Then I would love to see a demo that uses /dev/sg* with a ATAPI drive > using DMA for all related sector sizes. Use /dev/hdX or /dev/sdX or /dev/srX and SG_IO and it'll just work. Wonderful, huh? -- Jens Axboe