[SCSI] qla2xxx: Correct loop-resync issues during SNS scans.
authorAndrew Vasquez <andrew.vasquez@qlogic.com>
Thu, 18 Feb 2010 18:07:27 +0000 (10:07 -0800)
committerJames Bottomley <James.Bottomley@suse.de>
Fri, 19 Feb 2010 16:46:45 +0000 (10:46 -0600)
Original code could inadvertently skip the post processing of
port information returned from the SNS scan, if any link-flopping
asynchronous-events were received (noticed in FCAL topologies).

Cc: stable@kernel.org
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
drivers/scsi/qla2xxx/qla_init.c

index 1128c8d..96a609b 100644 (file)
@@ -2910,8 +2910,13 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha,
                if (qla2x00_is_reserved_id(vha, loop_id))
                        continue;
 
-               if (atomic_read(&vha->loop_down_timer) || LOOP_TRANSITION(vha))
+               if (atomic_read(&vha->loop_down_timer) ||
+                   LOOP_TRANSITION(vha)) {
+                       atomic_set(&vha->loop_down_timer, 0);
+                       set_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags);
+                       set_bit(LOCAL_LOOP_UPDATE, &vha->dpc_flags);
                        break;
+               }
 
                if (swl != NULL) {
                        if (last_dev) {