On 09/28/2016 10:46 PM, Hannes Reinecke wrote:
On 09/29/2016 01:45 AM, Bart Van Assche wrote:
The code at the end of alua_rtpg_work() is as follows:
scsi_device_put(sdev);
kref_put(&pg->kref, release_port_group);
Make sure that all code that queues the work item associated
with alua_rtpg_work() holds both references.
Signed-off-by: Bart Van Assche <bart.vanassche@xxxxxxxxxxx>
Cc: Hannes Reinecke <hare@xxxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
---
drivers/scsi/device_handler/scsi_dh_alua.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
index 752b5c9..f7099e2 100644
--- a/drivers/scsi/device_handler/scsi_dh_alua.c
+++ b/drivers/scsi/device_handler/scsi_dh_alua.c
@@ -889,6 +889,7 @@ static void alua_rtpg_queue(struct alua_port_group *pg,
/* Do not queue if the worker is already running */
if (!(pg->flags & ALUA_PG_RUNNING)) {
kref_get(&pg->kref);
+ scsi_device_get(sdev);
start_queue = 1;
}
}
Nack.
The scsi device reference tracks ->rtpg_sdev, not the queue item.
So we only need to take the scsi device reference if we're assigning a
new value to ->rtpg_sdev, which we are not.
Hello Hannes,
If queue_delayed_work() returns 0 that means that a call to
alua_rtpg_work() has already been scheduled. alua_rtpg_work() will drop
the reference on pg->rtpg_sdev. Are you sure that alua_rtpg_queue()
should call scsi_device_put(sdev) in that case?
Thanks,
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
