From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick Mansfield Subject: Re: [RFC] 3rd Party Device Drivers need sd.h Date: Fri, 17 Oct 2003 13:33:27 -0700 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <20031017133327.A1784@beaverton.ibm.com> References: <3F904E7F.1070700@comcast.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from e4.ny.us.ibm.com ([32.97.182.104]:59607 "EHLO e4.ny.us.ibm.com") by vger.kernel.org with ESMTP id S263510AbTJQUd5 (ORCPT ); Fri, 17 Oct 2003 16:33:57 -0400 Content-Disposition: inline In-Reply-To: <3F904E7F.1070700@comcast.net>; from david.vanhoose@comcast.net on Fri, Oct 17, 2003 at 04:18:07PM -0400 List-Id: linux-scsi@vger.kernel.org To: David van Hoose Cc: linux-scsi@vger.kernel.org On Fri, Oct 17, 2003 at 04:18:07PM -0400, David van Hoose wrote: > I have been working on modifying the SATA driver that I got from Promise > to work under 2.6.0. However, there is a conflict since it needs sd.h. I > looked and sd.h was merged into sd.c. What reasons were there for doing > this, and what changes have to be made to 3rd party drivers to > accomodate it? Please CC to me as I am not on this list. The bios_param call now passes a scsi_device and its capacity, rather than a scsi_disk, it is a cleaner interface. Exporting interfaces or data structures from sd to host drivers makes little sense. Just delete the sd.h include, and then updating the bios_param function should get rid of the scsi_disk or Disk reference. Does Jeff's SATA/libata driver cover the Promise hardware? -- Patrick Mansfield