[e1000 netdev-2.6 4/6] 82547 interrupt assert/de-assert re-ordering

["Feldman, Scott" <scott.feldman@xxxxxxxxx>]

* 82547 needs interrupt disable/enable to keep interrupt assertion
  state synced between 82547 and APIC.  82547 will re-order
  assert and de-assert messages if hub link bus is busy (heavy
  traffic).  Disabling interrupt on device works around re-
  order issue.  Note: this is a re-patch.  We backed out the
  patch because of a report on a system with a 8086:1019 device
  would lock up with this patch.  Turns out that system was a
  pre-production sample.

-----------

diff -Naurp netdev-2.6/drivers/net/e1000/e1000_main.c 
netdev-2.6/drivers/net/e1000.mod/e1000_main.c
--- netdev-2.6/drivers/net/e1000/e1000_main.c   2004-02-02 12:09:08.000000000 
-0800
+++ netdev-2.6/drivers/net/e1000.mod/e1000_main.c       2004-02-02 
12:10:43.000000000 -0800
@@ -2119,10 +2119,26 @@ e1000_intr(int irq, void *data, struct p
                __netif_rx_schedule(netdev);
        }
 #else
+        /* Writing IMC and IMS is needed for 82547.
+          Due to Hub Link bus being occupied, an interrupt 
+          de-assertion message is not able to be sent. 
+          When an interrupt assertion message is generated later,
+          two messages are re-ordered and sent out.
+          That causes APIC to think 82547 is in de-assertion
+          state, while 82547 is in assertion state, resulting 
+          in dead lock. Writing IMC forces 82547 into 
+          de-assertion state.
+        */
+       if(hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2)
+               e1000_irq_disable(adapter);
+
        for(i = 0; i < E1000_MAX_INTR; i++)
                if(!e1000_clean_rx_irq(adapter) &
                   !e1000_clean_tx_irq(adapter))
                        break;
+
+       if(hw->mac_type == e1000_82547 || hw->mac_type == e1000_82547_rev_2)
+               e1000_irq_enable(adapter);
 #endif
 
        return IRQ_HANDLED;