e1000e: tx_timeout should not increment for non-hang events
authorJesse Brandeburg <jesse.brandeburg@intel.com>
Wed, 2 Feb 2011 10:19:55 +0000 (10:19 +0000)
committerJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tue, 8 Feb 2011 09:10:17 +0000 (01:10 -0800)
Currently the driver increments the tx_timeout counter (an error counter)
when simply resetting the part with outstanding transmit work pending.
This is an unnecessary count of an error, when all we should be doing is
just resetting the part and discarding the transmits.  With this change the
only increment of tx_timeout is when the stack calls the watchdog reset
function due to a true Tx timeout.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Reviewed-by: Bruce Allan <bruce.w.allan@intel.com>
Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
drivers/net/e1000e/netdev.c

index 1c18f26b0812ac962708c68711abe58167c0eee0..3065870cf2a78871dff8b4856f15bf6ff58c113b 100644 (file)
@@ -4309,7 +4309,6 @@ link_up:
                         * to get done, so reset controller to flush Tx.
                         * (Do the reset outside of interrupt context).
                         */
-                       adapter->tx_timeout_count++;
                        schedule_work(&adapter->reset_task);
                        /* return immediately since reset is imminent */
                        return;