From mboxrd@z Thu Jan 1 00:00:00 1970 From: Douglas Gilbert Subject: [PATCH] scsi_reset_provider() in scsi_error.c lk2.5.52 Date: Tue, 24 Dec 2002 10:34:10 +1100 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <3E079D72.1010909@torque.net> Reply-To: dougg@torque.net Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="------------030106080700000608010103" Return-path: Received: from torque.net (dm1-32.triode.net.au [202.147.125.32]) by iggy.triode.net.au (8.11.6/8.11.6) with ESMTP id gBNNW8i18806 for ; Tue, 24 Dec 2002 10:32:09 +1100 List-Id: linux-scsi@vger.kernel.org To: linux-scsi@vger.kernel.org This is a multi-part message in MIME format. --------------030106080700000608010103 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit While testing Willem's last ide-scsi "rev1" patch with sg_reset I found that scsi_reset_provider() attempts to delete a timer that has not been set. A crude patch is attached that stops the kernel noise. Hopefully James can improve on it. Doug Gilbert --------------030106080700000608010103 Content-Type: text/plain; name="scsi_error_2552_reset.diff" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="scsi_error_2552_reset.diff" --- linux/drivers/scsi/scsi_error.c 2002-12-22 16:19:53.000000000 +1100 +++ linux/drivers/scsi/scsi_error.c2552res 2002-12-24 10:16:32.000000000 +1100 @@ -1811,6 +1811,10 @@ */ SCpnt->pid = 0; + /* Code expects a timer to be set, so set one for 120 seconds */ + init_timer(&SCpnt->eh_timeout); + scsi_add_timer(SCpnt, 120 * HZ, scsi_eh_times_out); + rtn = scsi_new_reset(SCpnt, flag); scsi_delete_timer(SCpnt); --------------030106080700000608010103--