From: Arnd Bergmann <arnd@arndb.de>
To: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Hannes Reinecke <hare@suse.com>, Arnd Bergmann <arnd@arndb.de>,
        Don Brace <don.brace@microsemi.com>,
        "James E.J. Bottomley" <jejb@linux.vnet.ibm.com>,
        Kevin Barnett <kevin.barnett@microsemi.com>,
        Scott Teel <scott.teel@microsemi.com>,
        Justin Lindley <justin.lindley@microsemi.com>,
        esc.storagedev@microsemi.com, linux-scsi@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: [PATCH] scsi: hpsa: fix uninitialized variable access
Date: Tue, 22 Nov 2016 15:32:54 +0100
Message-Id: <20161122143303.1871678-1-arnd@arndb.de>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1268
Lines: 40

A bugfix has left the 'sd' variable uninitialized:

drivers/scsi/hpsa.c: In function 'hpsa_slave_alloc':
drivers/scsi/hpsa.c:2033:5: error: 'sd' may be used uninitialized in this function [-Werror=maybe-uninitialized]

This reverts back to calling lookup_hpsa_scsi_dev() for the
HPSA_PHYSICAL_DEVICE_BUS case, but also keeps doing that when
hpsa_find_device_by_sas_rphy() returns NULL, as is currently
done.

The patch that caused this is marked for stable backports,
so this one has to be backported on top as well.

Fixes: 4eb307f7b18d ("scsi: hpsa: use bus '3' for legacy HBA devices")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
I did not try hard to figure out what the correct behavior
should be, so please treat this as a bugreport that might contain
the right fix.
---
 drivers/scsi/hpsa.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index ea64c01f3d42..d17ee63045c3 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -2029,7 +2029,10 @@ static int hpsa_slave_alloc(struct scsi_device *sdev)
 			sd->target = sdev_id(sdev);
 			sd->lun = sdev->lun;
 		}
+	} else {
+		sd = NULL;
 	}
+
 	if (!sd)
 		sd = lookup_hpsa_scsi_dev(h, sdev_channel(sdev),
 					sdev_id(sdev), sdev->lun);
-- 
2.9.0