From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: LDM scsi upper level updates and scsihosts parm Date: Sat, 31 May 2003 10:32:55 +0100 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <20030531103255.A28131@infradead.org> References: <20030531010844.GA25468@beaverton.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from phoenix.mvhi.com ([195.224.96.167]:6921 "EHLO phoenix.infradead.org") by vger.kernel.org with ESMTP id S264252AbTEaJTf (ORCPT ); Sat, 31 May 2003 05:19:35 -0400 Received: from hch by phoenix.infradead.org with local (Exim 4.10) id 19M2jP-0007Jm-00 for linux-scsi@vger.kernel.org; Sat, 31 May 2003 10:32:55 +0100 Content-Disposition: inline In-Reply-To: <20030531010844.GA25468@beaverton.ibm.com>; from andmike@us.ibm.com on Fri, May 30, 2003 at 06:08:45PM -0700 List-Id: linux-scsi@vger.kernel.org To: linux-scsi@vger.kernel.org On Fri, May 30, 2003 at 06:08:45PM -0700, Mike Anderson wrote: > The sysfs changes to allow the upper level to bind through the driver > model interfaces breaks the usage of the scsihosts parm. I did not > notice this the other day as I was running with one built-in driver and > many modules. > > I know there are issues with using scsihosts today (i.e., does not > work if you come in after the upper levels have inited). > > Are we going to remove the scsihosts parm or do we need a solution for > this? I think we should remove it. It's racy and we have proper ways to identify filesysystems bz label/uuid or scsi hosts by the sysfs output nowdays. Btw, did someone update the scsidev util to use 2.5 sysfs output?