net: dont update dev->trans_start in 10GB drivers
Followup of commits 9d21493b4beb8f918ba248032fefa393074a5e2b
and 08baf561083bc27a953aa087dd8a664bb2b88e8e
(net: tx scalability works : trans_start)
(net: txq_trans_update() helper)
Now that core network takes care of trans_start updates, dont do it
in drivers themselves, if possible. Multi queue drivers can
avoid one cache miss (on dev->trans_start) in their start_xmit()
handler.
Exceptions are NETIF_F_LLTX drivers (vxge & tehuti)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/drivers/net/vxge/vxge-main.c b/drivers/net/vxge/vxge-main.c
index b7f08f3..6c838b3 100644
--- a/drivers/net/vxge/vxge-main.c
+++ b/drivers/net/vxge/vxge-main.c
@@ -677,7 +677,7 @@
return VXGE_HW_OK;
}
-/* select a vpath to trasmit the packet */
+/* select a vpath to transmit the packet */
static u32 vxge_get_vpath_no(struct vxgedev *vdev, struct sk_buff *skb,
int *do_lock)
{
@@ -992,7 +992,9 @@
VXGE_HW_FIFO_TXD_TX_CKO_UDP_EN);
vxge_hw_fifo_txdl_post(fifo_hw, dtr);
- dev->trans_start = jiffies;
+#ifdef NETIF_F_LLTX
+ dev->trans_start = jiffies; /* NETIF_F_LLTX driver :( */
+#endif
spin_unlock_irqrestore(&fifo->tx_lock, flags);
VXGE_COMPLETE_VPATH_TX(fifo);