From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261595AbVD0Fe2 (ORCPT ); Wed, 27 Apr 2005 01:34:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261573AbVD0Fe1 (ORCPT ); Wed, 27 Apr 2005 01:34:27 -0400 Received: from stat16.steeleye.com ([209.192.50.48]:29862 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S261558AbVD0FeX (ORCPT ); Wed, 27 Apr 2005 01:34:23 -0400 Subject: Re: [PATCH scsi-misc-2.6 01/04] scsi: make scsi_send_eh_cmnd use its own timer instead of scmd->eh_timeout From: James Bottomley To: Tejun Heo Cc: SCSI Mailing List , Linux Kernel In-Reply-To: <426EF781.6040403@gmail.com> References: <20050419143100.E231523D@htj.dyndns.org> <20050419143100.0F9A8C3B@htj.dyndns.org> <1114381342.4786.17.camel@mulgrave> <426C2FC3.4090105@gmail.com> <1114452544.5000.11.camel@mulgrave> <426EF781.6040403@gmail.com> Content-Type: text/plain Date: Tue, 26 Apr 2005 22:34:19 -0700 Message-Id: <1114580059.5039.6.camel@mulgrave> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 (2.0.4-4) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2005-04-27 at 11:22 +0900, Tejun Heo wrote: > * A command is passed to lldd and starts execution > * It times out. > * eh runs > * abort isn't implemented or fails > * eh issues eh cmd (TUL, STU...) > * The command miraculously & stupidly completes just now. This should be impossible. The error handler API requirement is that the driver relinquish a command once it returns success from any error handling callback ... and if it never returns success, we simply offline the device and never use it again. This is the principle behind the command reuse: we only try an additional command *after* error handling succeeds, so the error handler now owns the command absolutely. > * The lldd succeeds to delete timer and normal completion path runs. > * We're fucked up now. James