From: Rob Evers <revers@redhat.com>
Date: Thu, 17 Jun 2010 16:44:15 -0400
Subject: [message] mptsas: fix disk add failing due to timeout
Message-id: <20100617164058.7375.81599.sendpatchset@localhost.localdomain>
Patchwork-id: 26285
O-Subject: [RHEL5.6 PATCH] mptsas: The addition of SAS disk fails because of the
	timeout
Bugzilla: 542892
RH-Acked-by: Tomas Henzl <thenzl@redhat.com>

https://bugzilla.redhat.com/show_bug.cgi?id=542892

This patch addresses a case where a /dev/sd is deleted via
'echo 1 > /sys/class/scsi_device/a:b:c:d/device/delete'
and a subsequent power cycle of the disk results in the disk
not being reconfigured.

The patch is provided by LSI

The patch was tested by LSI and confirmed as functioning correctly
in the reported failing case, and the case of simply powering the disk
off and on, and seeing the disk be reconfigured as expected in both
cases.

Signed-off-by: Jarod Wilson <jarod@redhat.com>

diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c
index 4b2a2f3..c620283 100644
--- a/drivers/message/fusion/mptsas.c
+++ b/drivers/message/fusion/mptsas.c
@@ -1098,10 +1098,8 @@ mptsas_target_reset_queue(MPT_ADAPTER *ioc,
 	id = sas_event_data->TargetID;
 	channel = sas_event_data->Bus;
 
-	if (!(vtarget = mptsas_find_vtarget(ioc, channel, id)))
-		return;
-
-	vtarget->deleted = 1; /* block IO */
+	if ((vtarget = mptsas_find_vtarget(ioc, channel, id)))
+		vtarget->deleted = 1; /* block IO */
 
 	target_reset_list = kzalloc(sizeof(struct mptsas_target_reset_event),
 	    GFP_ATOMIC);
@@ -1207,7 +1205,7 @@ mptsas_taskmgmt_complete(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *mr)
 	 * enable work queue to remove device from upper layers
 	 */
 	list_del(&target_reset_list->list);
-	if ((mptsas_find_vtarget(ioc, channel, id)) && !ioc->fw_events_off)
+	if (!ioc->fw_events_off)
 		mptsas_queue_device_delete(ioc,
 			&target_reset_list->sas_event_data);