From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick Mansfield Subject: Re: [PATCH 1/1] ipr: Fix for adapter shutdown issue Date: Wed, 15 Jun 2005 10:05:45 -0700 Message-ID: <20050615170545.GB23912@us.ibm.com> References: <200506151515.j5FFFNQA020480@d01av01.pok.ibm.com> <1118849292.5045.27.camel@mulgrave> <42B04A9C.7050503@us.ibm.com> <1118850458.5045.34.camel@mulgrave> <42B05330.7050406@us.ibm.com> <1118853899.5045.39.camel@mulgrave> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from e31.co.us.ibm.com ([32.97.110.129]:19186 "EHLO e31.co.us.ibm.com") by vger.kernel.org with ESMTP id S261220AbVFORG3 (ORCPT ); Wed, 15 Jun 2005 13:06:29 -0400 Received: from d03relay04.boulder.ibm.com (d03relay04.boulder.ibm.com [9.17.195.106]) by e31.co.us.ibm.com (8.12.10/8.12.9) with ESMTP id j5FH6Sua442002 for ; Wed, 15 Jun 2005 13:06:28 -0400 Received: from d03av02.boulder.ibm.com (d03av02.boulder.ibm.com [9.17.195.168]) by d03relay04.boulder.ibm.com (8.12.10/NCO/VER6.6) with ESMTP id j5FH6S0E172352 for ; Wed, 15 Jun 2005 11:06:28 -0600 Received: from d03av02.boulder.ibm.com (loopback [127.0.0.1]) by d03av02.boulder.ibm.com (8.12.11/8.13.3) with ESMTP id j5FH6RmQ032258 for ; Wed, 15 Jun 2005 11:06:27 -0600 Content-Disposition: inline In-Reply-To: <1118853899.5045.39.camel@mulgrave> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: James Bottomley Cc: brking@us.ibm.com, SCSI Mailing List , haren@us.ibm.com On Wed, Jun 15, 2005 at 11:44:59AM -0500, James Bottomley wrote: > > Well, I still think you're operating unsafely ... some people let these > batteries run out, you know ... > > To fix this, add a > > blk_queue_ordered(sdev->request_queue, QUEUE_ORDERED_NONE); > > to the slave configure routine for each of the devices that doesn't > synchronize the cache. That should correct the SCSI layer assumptions. I don't get it. First in scsi_alloc_queue, the flush/sync code is only used if we don't have ordered tags, and if the adapter driver explicitly allows it (shost->order_flush). Plus we clear order_flush in scsi_host_alloc if can_queue > 1. So ipr will never get those sync cache commands, correct? And we default to QUEUE_ORDERED_NONE (0) if ordered_tag and ordered_flush are not set. But I can't find any where that we set ordered_tag at all, am I missing something??? Let alone in ipr. -- Patrick Mansfield