From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [patch] scsi/libata: correct bug for ULi M5281 Date: Sat, 05 Feb 2005 23:43:33 -0500 Message-ID: <4205A075.8090405@pobox.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:45744 "EHLO parcelfarce.linux.theplanet.co.uk") by vger.kernel.org with ESMTP id S268914AbVBFEnu (ORCPT ); Sat, 5 Feb 2005 23:43:50 -0500 In-Reply-To: Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Clear.Zhang@uli.com.tw Cc: linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org, andrebalsa@mailingaddress.org, Peer.Chen@uli.com.tw, Emily.Jiang@uli.com.tw, Eric.Lo@uli.com.tw Clear.Zhang@uli.com.tw wrote: > Hi Jeff, > > I've correct the patch according to your suggestion. > But we need libata-core.c to export some functions. > Here is the patch. Please check it and apply it to new kernels. > Thanks a lot. I'm afraid I cannot accept even the updated version of your patch. Your second patch updates libata-core, and adds ULi-specific code to it. What we need to do is 1) determine if your problem is ULi-specific, or applies to other controllers 2a) If the problem is not ULi-specific, simply update libata to new behavior. 2b) If the problem is ULi-specific, libata must be modified such that all ULi-specific reset procedures are in sata_uli.c, and _no code is duplicated from libata_. This may require that we change the libata API. In Linux, it is OK to change the driver API when there is a problem :) Regards, Jeff