i40e: fix TCP flag replication for hardware offload

As reported by Eric Dumazet, the i40e driver was allowing the hardware
to replicate the PSH flag on all segments of a TSO operation.

This patch fixes the first/middle/last TCP flags settings which
makes the TSO operations work correctly.

With this change we are now configuring the CWR bit to only be set
in the first packet of a TSO, so this patch also enables TSO_ECN,
in order to advertise to the stack that we do the right thing
on the wire.

Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 861b722..59eada3 100644 (file)
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -4271,6 +4271,14 @@ static int i40e_open(struct net_device *netdev)
        if (err)
                return err;
 
+       /* configure global TSO hardware offload settings */
+       wr32(&pf->hw, I40E_GLLAN_TSOMSK_F, be32_to_cpu(TCP_FLAG_PSH |
+                                                      TCP_FLAG_FIN) >> 16);
+       wr32(&pf->hw, I40E_GLLAN_TSOMSK_M, be32_to_cpu(TCP_FLAG_PSH |
+                                                      TCP_FLAG_FIN |
+                                                      TCP_FLAG_CWR) >> 16);
+       wr32(&pf->hw, I40E_GLLAN_TSOMSK_L, be32_to_cpu(TCP_FLAG_CWR) >> 16);
+
 #ifdef CONFIG_I40E_VXLAN
        vxlan_get_rx_port(netdev);
 #endif
@@ -6712,6 +6720,7 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
                           NETIF_F_HW_VLAN_CTAG_FILTER |
                           NETIF_F_IPV6_CSUM           |
                           NETIF_F_TSO                 |
+                          NETIF_F_TSO_ECN             |
                           NETIF_F_TSO6                |
                           NETIF_F_RXCSUM              |
                           NETIF_F_NTUPLE              |