From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: RE: [PATCH] fix for BZ 8426 - massive slowdown on SCSI CD/DVDdriveconnected to mptspi driver Date: Thu, 31 May 2007 11:14:11 -0500 Message-ID: <1180628051.3743.34.camel@mulgrave.il.steeleye.com> References: <664A4EBB07F29743873A87CF62C26D707881DB@NAMAIL4.ad.lsil.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from hancock.steeleye.com ([71.30.118.248]:33565 "EHLO hancock.sc.steeleye.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752826AbXEaQON (ORCPT ); Thu, 31 May 2007 12:14:13 -0400 In-Reply-To: <664A4EBB07F29743873A87CF62C26D707881DB@NAMAIL4.ad.lsil.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: "Moore, Eric" Cc: linux-scsi@vger.kernel.org, Doug Chapman On Thu, 2007-05-31 at 09:40 -0600, Moore, Eric wrote: > James - I obtained a log from Doug (attached), and the firmware is > returning BUSY, so your correct. That will end up as a ADD_TO_MLQUEUE > disposition. Does the mid-layer throttle back the queue when a device > returns busy, I don't know? I still believe that the lld shouldn't > report that a device supports tagged when it really doesn't. I hope > you will consider this patch, if not, let me know how to proceed. Sorry, yes ... I should have said that the patch is correct and I'll put it in 2.6.22 However, since we have such a wonderful example of the failing case, I'd like to try to correct the root cause, if I may. Doug, would you be amenable to testing out a few patches for me? James