From mboxrd@z Thu Jan 1 00:00:00 1970 From: Randy Dunlap Subject: Re: [PATCH] lpfc : add module parameter that allows adapter instances to avoid attachment Date: Thu, 26 Mar 2009 13:41:15 -0700 Message-ID: <49CBE86B.4050701@oracle.com> References: <1238097300.27023.2.camel@ogier> <1238098747.3342.54.camel@localhost.localdomain> <49CBE772.80304@emulex.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from acsinet12.oracle.com ([141.146.126.234]:60665 "EHLO acsinet12.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753677AbZCZUlv (ORCPT ); Thu, 26 Mar 2009 16:41:51 -0400 In-Reply-To: <49CBE772.80304@emulex.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: James Smart Cc: James Bottomley , "linux-scsi@vger.kernel.org" James Smart wrote: > The oem requirement we have explicitly states not to use the unbind > interface > (we proposed unbind at first as well). > > The issue is what happens on the link while we are bound for that short > amount of time. It confuses the things on the other side of the link. > There's a secondary driver that ends up binding to the adapters we exclude, > and the things on the other side only expected to see the second driver. > > -- james s > > > James Bottomley wrote: >> On Thu, 2009-03-26 at 15:55 -0400, James Smart wrote: >>> This patch adds a module parameter that supplies a text string >>> containing a >>> list of PCI :. values to identify adapter instances >>> that >>> should *not* be attached to by the driver. What does this do on systems that use PCI domains? domain:bus:slot.func (??) >>> -- james s >>> >>> >>> Signed-off-by: James Smart >> >> Is there a reason why you can't just use the generic unbind interface? >> >> James -- ~Randy