From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: LibATA ATAPI-Found lockup spot. Date: Tue, 18 Jan 2005 14:43:20 -0500 Message-ID: <41ED66D8.2040808@pobox.com> References: <20050118191122.4DC7ACA0AF@ws7-4.us4.outblaze.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:13224 "EHLO parcelfarce.linux.theplanet.co.uk") by vger.kernel.org with ESMTP id S261452AbVARTnn (ORCPT ); Tue, 18 Jan 2005 14:43:43 -0500 In-Reply-To: <20050118191122.4DC7ACA0AF@ws7-4.us4.outblaze.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Kien Pham Cc: "linux-ide@vger.kernel.org" Kien Pham wrote: > Hi Jeff, > > It blocks forever on line 2450 of libata-core.c. This is on vanilla > 2.6.10 kernel source. Relevent snippet follows so you don't have to dig. > Basically, the middle SCSI layer successfully sends 3 INQUIRY commands. > Then it sends a REPORT LUNs command that fails with a status 0x50. The > error register reads 0x60 at this point. Then the ATAPI layer sends a > REQUEST SENSE which never completes because an interrupt never occurs > and we're stuck at line 2450 waiting for completion. O.K. I know the how > it happens, any ideas on the why? Or am I on my own? :) I think Bart mentioned something about problems with REPORT LUNS... I bet things would work if we didn't send that command... Jeff