From mboxrd@z Thu Jan 1 00:00:00 1970 From: Luben Tuikov Subject: Re: no INQUIRY from userspace please (was Re: [PATCH 7/9] scsi_dh: Add support for SDEV_PASSIVE) Date: Thu, 7 Feb 2008 12:42:31 -0800 (PST) Message-ID: <157925.90863.qm@web31814.mail.mud.yahoo.com> References: <1202396490.3171.8.camel@localhost.localdomain> Reply-To: ltuikov@yahoo.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from web31814.mail.mud.yahoo.com ([68.142.206.167]:25620 "HELO web31814.mail.mud.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1755327AbYBGUmg (ORCPT ); Thu, 7 Feb 2008 15:42:36 -0500 In-Reply-To: <1202396490.3171.8.camel@localhost.localdomain> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Stefan Richter , James Bottomley Cc: Mike Anderson , Mike Christie , sekharan@us.ibm.com, dm-devel@redhat.com, linux-scsi@vger.kernel.org, jens.axboe@oracle.com --- On Thu, 2/7/08, James Bottomley wrote: > This is all a tradeoff. If you want userspace *never* to > issue raw SCSI > commands like INQUIRY, we're going to have to provide > the needed > information from the kernel via sysfs ... including VPD > strings. This > is something we've always shovelled off into userspace > before. What if a user-space application client _does_ send an INQUIRY to a device anyway? It would probably be better to preserve application client behaviour and simulate/emulate n-th INQUIRY, after the 1st for such broken device firmwares that break on any subsequent INQUIRY. Possibly in the LLDD or via blacklisting in the mid-layer. Luben