Currently driver completes double completed or spurious interrupted fibs.
This  is not necessary and causes the SCSI mid layer to issue aborts and
resets, since completing a fib  prematurely might trigger a race condition
resulting in the driver not calling the scsi_done callback.
Fixed by removing the call to fib complete.
Signed-off-by: Raghava Aditya Renukunta <RaghavaAditya.Renukunta@microsemi.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
                        if (likely(fib->callback && fib->callback_data)) {
                                fib->flags &= FIB_CONTEXT_FLAG_FASTRESP;
                                fib->callback(fib->callback_data, fib);
-                       } else {
-                               aac_fib_complete(fib);
-                       }
+                       } else
+                               dev_info(&dev->pdev->dev,
+                               "Invalid callback_fib[%d] (*%p)(%p)\n",
+                               index, fib->callback, fib->callback_data);
                } else {
                        unsigned long flagv;
                        dprintk((KERN_INFO "event_wait up\n"));