[net-next 00/15][pull request] 40GbE Intel Wired LAN Driver Updates 2016-04-26

[net-next 00/15][pull request] 40GbE Intel Wired LAN Driver Updates 2016-04-26

[Jeff Kirsher]

This series contains updates to i40e and i40evf.

Alex Duyck cleans up the feature flags since they are becoming pretty
"massive", the primary change being that we now build our features list
around hw_encap_features.  Added support for IPIP and SIT offloads,
which should improvement in throughput for IPIP and SIT tunnels with
the offload enabled.

Mitch adds support for configuring RSS on behalf of the VFs, which removes
the burden of dealing with different hardware interfaces from the VF
drivers and improves future compatibility.  Fix to ensure that we do not
panic by checking that the vsi_res pointer is valid before dereferencing
it, after which we can drink beer and eat peanuts.

Shannon does come housekeeping in i40e_add_fdir_ethtool() in preparation
for more cloud filter work.  Added flexibility to the nvmupdate
facility by adding the ability to specify an AQ event opcode to wait on
after Exec_AQ request.

Kiran implements the feature to allow users to change input set mask for
flow director using side-band channel.

Michal adds device capability which defines if an update is available and
if a security check is needed during the update process.

Kamil just adds a device id to support X722 QSFP+ device.

Greg fixes an issue where a mirror rule ID may be zero, so do not return
invalid parameter when the user passes in a zero for a rule ID.  Adds
support to steer packets to VSIs by VLAN tag alone while being in
promiscuous mode for multicast and unicast MAC addresses.

Jesse fixes the driver from offloading the VLAN tag into the skb any
time there was a VLAN tag and the hardware stripping was enabled, to
making sure it is enabled before put_tag.

The following are changes since commit fab7b629a82da1b59620470d13152aff975239f6:
  Merge branch 'ila-csum-neutral'
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 40GbE

Alexander Duyck (2):
  i40e/i40evf: Clean up feature flags
  i40e/i40evf: Add support for IPIP and SIT offloads

Anjali Singhai Jain (2):
  i40e: Add VF promiscuous mode driver support
  i40evf: Add driver support for promiscuous mode

Greg Rose (2):
  i40e: Remove zero check
  i40e: Add promiscuous on VLAN support

Jesse Brandeburg (1):
  i40e/i40evf: Only offload VLAN tag if enabled

Kamil Krawczyk (1):
  i40e: Add DeviceID for X722 QSFP+

Kiran Patil (1):
  i40e: Allow user to change input set mask for flow director

Michal Kosiarz (1):
  i40e: Add device capability which defines if update is available

Mitch Williams (3):
  i40e: Add support for configuring VF RSS
  i40evf: Don't Panic
  i40evf: Allow PF driver to configure RSS

Shannon Nelson (2):
  i40e: Code cleanup in i40e_add_fdir_ethtool
  i40e: Specify AQ event opcode to wait for

 drivers/net/ethernet/intel/i40e/i40e.h             |   7 +
 drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h  |   1 +
 drivers/net/ethernet/intel/i40e/i40e_common.c      |  82 ++++-
 drivers/net/ethernet/intel/i40e/i40e_devids.h      |   1 +
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c     | 128 ++++++-
 drivers/net/ethernet/intel/i40e/i40e_main.c        |  98 ++---
 drivers/net/ethernet/intel/i40e/i40e_nvm.c         |  49 ++-
 drivers/net/ethernet/intel/i40e/i40e_prototype.h   |   8 +
 drivers/net/ethernet/intel/i40e/i40e_txrx.c        |  29 +-
 drivers/net/ethernet/intel/i40e/i40e_type.h        |   6 +
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 346 +++++++++++++++++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |   2 +
 .../net/ethernet/intel/i40evf/i40e_adminq_cmd.h    |   1 +
 drivers/net/ethernet/intel/i40evf/i40e_common.c    |   1 +
 drivers/net/ethernet/intel/i40evf/i40e_devids.h    |   1 +
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c      |  27 +-
 drivers/net/ethernet/intel/i40evf/i40e_type.h      |   6 +
 drivers/net/ethernet/intel/i40evf/i40evf.h         |  39 +-
 drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c | 121 +++----
 drivers/net/ethernet/intel/i40evf/i40evf_main.c    | 395 ++++++++-------------
 .../net/ethernet/intel/i40evf/i40evf_virtchnl.c    | 130 +++++++
 21 files changed, 1070 insertions(+), 408 deletions(-)

-- 
2.5.5

[net-next 01/15] i40e/i40evf: Clean up feature flags

[Jeff Kirsher]

From: Alexander Duyck <aduyck@mirantis.com>

The feature flags list for i40e and i40evf is beginning to become pretty
massive.  I plan to add another 4 or so features to these drivers and
duplicating the flags for each and every flags list is becoming a bit
repetitive.

The primary change here is that we now build our features list around
hw_encap_features.  After that we assign that to vlan_features,
hw_features, and finally map that onto features.  In addition we end up
throwing features onto hw_encap_features that end up having no effect such
as the Rx offloads and SCTP_CRC.  However that should have no impact and
makes things a bit easier for us as hw_encap_features is one of the less
updated features maps available.

For i40evf I went through and sanity checked a few features as well.
Specifically RXCSUM was being set as a read-only feature which didn't make
much sense.  I have updated things so we can clear the NETIF_F_RXCSUM flag
since that is really a software feature and not a hardware one anyway so
disabling it is just a matter of ignoring the result from the hardware.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c     | 61 ++++++++++-------------
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 66 ++++++++++++-------------
 2 files changed, 58 insertions(+), 69 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 0b071ce..f2e83fe 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -9111,40 +9111,36 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 	np = netdev_priv(netdev);
 	np->vsi = vsi;
 
-	netdev->hw_enc_features |= NETIF_F_IP_CSUM	       |
-				   NETIF_F_IPV6_CSUM	       |
-				   NETIF_F_TSO		       |
-				   NETIF_F_TSO6		       |
-				   NETIF_F_TSO_ECN	       |
-				   NETIF_F_GSO_GRE	       |
-				   NETIF_F_GSO_UDP_TUNNEL      |
-				   NETIF_F_GSO_UDP_TUNNEL_CSUM |
+	netdev->hw_enc_features |= NETIF_F_SG			|
+				   NETIF_F_IP_CSUM		|
+				   NETIF_F_IPV6_CSUM		|
+				   NETIF_F_HIGHDMA		|
+				   NETIF_F_SOFT_FEATURES	|
+				   NETIF_F_TSO			|
+				   NETIF_F_TSO_ECN		|
+				   NETIF_F_TSO6			|
+				   NETIF_F_GSO_GRE		|
+				   NETIF_F_GSO_UDP_TUNNEL	|
+				   NETIF_F_GSO_UDP_TUNNEL_CSUM	|
+				   NETIF_F_SCTP_CRC		|
+				   NETIF_F_RXHASH		|
+				   NETIF_F_RXCSUM		|
 				   0;
 
-	netdev->features = NETIF_F_SG		       |
-			   NETIF_F_IP_CSUM	       |
-			   NETIF_F_SCTP_CRC	       |
-			   NETIF_F_HIGHDMA	       |
-			   NETIF_F_GSO_UDP_TUNNEL      |
-			   NETIF_F_GSO_GRE	       |
-			   NETIF_F_HW_VLAN_CTAG_TX     |
-			   NETIF_F_HW_VLAN_CTAG_RX     |
-			   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_RXHASH	       |
-			   0;
+	if (!(pf->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE))
+		netdev->hw_enc_features ^= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+	/* record features VLANs can make use of */
+	netdev->vlan_features |= netdev->hw_enc_features;
 
 	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
-		netdev->features |= NETIF_F_NTUPLE;
-	if (pf->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE)
-		netdev->features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev->hw_features |= NETIF_F_NTUPLE;
+
+	netdev->hw_features |= netdev->hw_enc_features	|
+			       NETIF_F_HW_VLAN_CTAG_TX	|
+			       NETIF_F_HW_VLAN_CTAG_RX;
 
-	/* copy netdev features into list of user selectable features */
-	netdev->hw_features |= netdev->features;
+	netdev->features |= netdev->hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
 
 	if (vsi->type == I40E_VSI_MAIN) {
 		SET_NETDEV_DEV(netdev, &pf->pdev->dev);
@@ -9183,12 +9179,7 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 
 	ether_addr_copy(netdev->dev_addr, mac_addr);
 	ether_addr_copy(netdev->perm_addr, mac_addr);
-	/* vlan gets same features (except vlan offload)
-	 * after any tweaks for specific VSI types
-	 */
-	netdev->vlan_features = netdev->features & ~(NETIF_F_HW_VLAN_CTAG_TX |
-						     NETIF_F_HW_VLAN_CTAG_RX |
-						   NETIF_F_HW_VLAN_CTAG_FILTER);
+
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 	netdev->priv_flags |= IFF_SUPP_NOFCS;
 	/* Setup netdev TC information */
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 9110319..e3857d8 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2337,40 +2337,38 @@ int i40evf_process_config(struct i40evf_adapter *adapter)
 		return -ENODEV;
 	}
 
-	netdev->features |= NETIF_F_HIGHDMA |
-			    NETIF_F_SG |
-			    NETIF_F_IP_CSUM |
-			    NETIF_F_SCTP_CRC |
-			    NETIF_F_IPV6_CSUM |
-			    NETIF_F_TSO |
-			    NETIF_F_TSO6 |
-			    NETIF_F_TSO_ECN |
-			    NETIF_F_GSO_GRE |
-			    NETIF_F_GSO_UDP_TUNNEL |
-			    NETIF_F_RXCSUM |
-			    NETIF_F_GRO;
-
-	netdev->hw_enc_features |= NETIF_F_IP_CSUM	       |
-				   NETIF_F_IPV6_CSUM	       |
-				   NETIF_F_TSO		       |
-				   NETIF_F_TSO6		       |
-				   NETIF_F_TSO_ECN	       |
-				   NETIF_F_GSO_GRE	       |
-				   NETIF_F_GSO_UDP_TUNNEL      |
-				   NETIF_F_GSO_UDP_TUNNEL_CSUM;
-
-	if (adapter->flags & I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE)
-		netdev->features |= NETIF_F_GSO_UDP_TUNNEL_CSUM;
-
-	/* always clear VLAN features because they can change at every reset */
-	netdev->features &= ~(I40EVF_VLAN_FEATURES);
-	/* copy netdev features into list of user selectable features */
-	netdev->hw_features |= netdev->features;
-
-	if (vfres->vf_offload_flags & I40E_VIRTCHNL_VF_OFFLOAD_VLAN) {
-		netdev->vlan_features = netdev->features;
-		netdev->features |= I40EVF_VLAN_FEATURES;
-	}
+	netdev->hw_enc_features |= NETIF_F_SG			|
+				   NETIF_F_IP_CSUM		|
+				   NETIF_F_IPV6_CSUM		|
+				   NETIF_F_HIGHDMA		|
+				   NETIF_F_SOFT_FEATURES	|
+				   NETIF_F_TSO			|
+				   NETIF_F_TSO_ECN		|
+				   NETIF_F_TSO6			|
+				   NETIF_F_GSO_GRE		|
+				   NETIF_F_GSO_UDP_TUNNEL	|
+				   NETIF_F_GSO_UDP_TUNNEL_CSUM	|
+				   NETIF_F_SCTP_CRC		|
+				   NETIF_F_RXHASH		|
+				   NETIF_F_RXCSUM		|
+				   0;
+
+	if (!(adapter->flags & I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE))
+		netdev->hw_enc_features ^= NETIF_F_GSO_UDP_TUNNEL_CSUM;
+
+	/* record features VLANs can make use of */
+	netdev->vlan_features |= netdev->hw_enc_features;
+
+	/* Write features and hw_features separately to avoid polluting
+	 * with, or dropping, features that are set when we registgered.
+	 */
+	netdev->hw_features |= netdev->hw_enc_features;
+
+	netdev->features |= netdev->hw_enc_features | I40EVF_VLAN_FEATURES;
+
+	/* disable VLAN features if not supported */
+	if (!(vfres->vf_offload_flags & I40E_VIRTCHNL_VF_OFFLOAD_VLAN))
+		netdev->features ^= I40EVF_VLAN_FEATURES;
 
 	adapter->vsi.id = adapter->vsi_res->vsi_id;
 
-- 
2.5.5

[net-next 02/15] i40e/i40evf: Add support for IPIP and SIT offloads

[Jeff Kirsher]

From: Alexander Duyck <aduyck@mirantis.com>

Looking over the documentation it turns out enabling IPIP and SIT offloads
for i40e is pretty straightforward.  As such I decided to enable them with
this patch.  In my testing I am seeing an improvement of 8 to 10 Gb/s
for IPIP and SIT tunnels with this offload enabled.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c     |  2 ++
 drivers/net/ethernet/intel/i40e/i40e_txrx.c     | 24 ++++++++++++++++--------
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c   | 24 ++++++++++++++++--------
 drivers/net/ethernet/intel/i40evf/i40evf_main.c |  2 ++
 4 files changed, 36 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f2e83fe..ec94ad6c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -9120,6 +9120,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
 				   NETIF_F_TSO_ECN		|
 				   NETIF_F_TSO6			|
 				   NETIF_F_GSO_GRE		|
+				   NETIF_F_GSO_IPIP		|
+				   NETIF_F_GSO_SIT		|
 				   NETIF_F_GSO_UDP_TUNNEL	|
 				   NETIF_F_GSO_UDP_TUNNEL_CSUM	|
 				   NETIF_F_SCTP_CRC		|
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 39efba0..6e44cf1 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -2299,7 +2299,10 @@ static int i40e_tso(struct sk_buff *skb, u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 		ip.v6->payload_len = 0;
 	}
 
-	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE |
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE |
+					 SKB_GSO_IPIP |
+					 SKB_GSO_SIT |
+					 SKB_GSO_UDP_TUNNEL |
 					 SKB_GSO_UDP_TUNNEL_CSUM)) {
 		if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
 			/* determine offset of outer transport header */
@@ -2442,13 +2445,6 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 						 &l4_proto, &frag_off);
 		}
 
-		/* compute outer L3 header size */
-		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
-			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
-
-		/* switch IP header pointer from outer to inner header */
-		ip.hdr = skb_inner_network_header(skb);
-
 		/* define outer transport */
 		switch (l4_proto) {
 		case IPPROTO_UDP:
@@ -2459,6 +2455,11 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			tunnel |= I40E_TXD_CTX_GRE_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
 			break;
+		case IPPROTO_IPIP:
+		case IPPROTO_IPV6:
+			*tx_flags |= I40E_TX_FLAGS_UDP_TUNNEL;
+			l4.hdr = skb_inner_network_header(skb);
+			break;
 		default:
 			if (*tx_flags & I40E_TX_FLAGS_TSO)
 				return -1;
@@ -2467,6 +2468,13 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			return 0;
 		}
 
+		/* compute outer L3 header size */
+		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
+			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
+
+		/* switch IP header pointer from outer to inner header */
+		ip.hdr = skb_inner_network_header(skb);
+
 		/* compute tunnel header size */
 		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
 			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index fc22818..f101895 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1564,7 +1564,10 @@ static int i40e_tso(struct sk_buff *skb, u8 *hdr_len, u64 *cd_type_cmd_tso_mss)
 		ip.v6->payload_len = 0;
 	}
 
-	if (skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_GRE |
+	if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE |
+					 SKB_GSO_IPIP |
+					 SKB_GSO_SIT |
+					 SKB_GSO_UDP_TUNNEL |
 					 SKB_GSO_UDP_TUNNEL_CSUM)) {
 		if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM) {
 			/* determine offset of outer transport header */
@@ -1665,13 +1668,6 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 						 &l4_proto, &frag_off);
 		}
 
-		/* compute outer L3 header size */
-		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
-			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
-
-		/* switch IP header pointer from outer to inner header */
-		ip.hdr = skb_inner_network_header(skb);
-
 		/* define outer transport */
 		switch (l4_proto) {
 		case IPPROTO_UDP:
@@ -1682,6 +1678,11 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			tunnel |= I40E_TXD_CTX_GRE_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
+		case IPPROTO_IPIP:
+		case IPPROTO_IPV6:
+			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
+			l4.hdr = skb_inner_network_header(skb);
+			break;
 		default:
 			if (*tx_flags & I40E_TX_FLAGS_TSO)
 				return -1;
@@ -1690,6 +1691,13 @@ static int i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			return 0;
 		}
 
+		/* compute outer L3 header size */
+		tunnel |= ((l4.hdr - ip.hdr) / 4) <<
+			  I40E_TXD_CTX_QW0_EXT_IPLEN_SHIFT;
+
+		/* switch IP header pointer from outer to inner header */
+		ip.hdr = skb_inner_network_header(skb);
+
 		/* compute tunnel header size */
 		tunnel |= ((ip.hdr - l4.hdr) / 2) <<
 			  I40E_TXD_CTX_QW0_NATLEN_SHIFT;
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index e3857d8..806da26 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2346,6 +2346,8 @@ int i40evf_process_config(struct i40evf_adapter *adapter)
 				   NETIF_F_TSO_ECN		|
 				   NETIF_F_TSO6			|
 				   NETIF_F_GSO_GRE		|
+				   NETIF_F_GSO_IPIP		|
+				   NETIF_F_GSO_SIT		|
 				   NETIF_F_GSO_UDP_TUNNEL	|
 				   NETIF_F_GSO_UDP_TUNNEL_CSUM	|
 				   NETIF_F_SCTP_CRC		|
-- 
2.5.5

[net-next 03/15] i40e: Add support for configuring VF RSS

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Add support for configuring RSS on behalf of the VFs. This removes the
burden of dealing with different hardware interfaces from the VF
drivers, allowing for better future compatibility.

Change-ID: Icea75d3f37241ee8e447be5779e5abb53ddf04c0
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h             |   1 +
 drivers/net/ethernet/intel/i40e/i40e_main.c        |  35 +++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 193 ++++++++++++++++++++-
 3 files changed, 217 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index d25b3be..e312adf 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -202,6 +202,7 @@ struct i40e_lump_tracking {
 
 #define I40E_HKEY_ARRAY_SIZE ((I40E_PFQF_HKEY_MAX_INDEX + 1) * 4)
 #define I40E_HLUT_ARRAY_SIZE ((I40E_PFQF_HLUT_MAX_INDEX + 1) * 4)
+#define I40E_VF_HLUT_ARRAY_SIZE ((I40E_VFQF_HLUT1_MAX_INDEX + 1) * 4)
 
 enum i40e_fd_stat_idx {
 	I40E_FD_STAT_ATR,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index ec94ad6c..39b3b56 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -8082,24 +8082,45 @@ static int i40e_config_rss_reg(struct i40e_vsi *vsi, const u8 *seed,
 {
 	struct i40e_pf *pf = vsi->back;
 	struct i40e_hw *hw = &pf->hw;
+	u16 vf_id = vsi->vf_id;
 	u8 i;
 
 	/* Fill out hash function seed */
 	if (seed) {
 		u32 *seed_dw = (u32 *)seed;
 
-		for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
-			i40e_write_rx_ctl(hw, I40E_PFQF_HKEY(i), seed_dw[i]);
+		if (vsi->type == I40E_VSI_MAIN) {
+			for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
+				i40e_write_rx_ctl(hw, I40E_PFQF_HKEY(i),
+						  seed_dw[i]);
+		} else if (vsi->type == I40E_VSI_SRIOV) {
+			for (i = 0; i <= I40E_VFQF_HKEY1_MAX_INDEX; i++)
+				i40e_write_rx_ctl(hw,
+						  I40E_VFQF_HKEY1(i, vf_id),
+						  seed_dw[i]);
+		} else {
+			dev_err(&pf->pdev->dev, "Cannot set RSS seed - invalid VSI type\n");
+		}
 	}
 
 	if (lut) {
 		u32 *lut_dw = (u32 *)lut;
 
-		if (lut_size != I40E_HLUT_ARRAY_SIZE)
-			return -EINVAL;
-
-		for (i = 0; i <= I40E_PFQF_HLUT_MAX_INDEX; i++)
-			wr32(hw, I40E_PFQF_HLUT(i), lut_dw[i]);
+		if (vsi->type == I40E_VSI_MAIN) {
+			if (lut_size != I40E_HLUT_ARRAY_SIZE)
+				return -EINVAL;
+			for (i = 0; i <= I40E_PFQF_HLUT_MAX_INDEX; i++)
+				wr32(hw, I40E_PFQF_HLUT(i), lut_dw[i]);
+		} else if (vsi->type == I40E_VSI_SRIOV) {
+			if (lut_size != I40E_VF_HLUT_ARRAY_SIZE)
+				return -EINVAL;
+			for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++)
+				i40e_write_rx_ctl(hw,
+						  I40E_VFQF_HLUT1(i, vf_id),
+						  lut_dw[i]);
+		} else {
+			dev_err(&pf->pdev->dev, "Cannot set RSS LUT - invalid VSI type\n");
+		}
 	}
 	i40e_flush(hw);
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 30f8cbe..c364588 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1348,12 +1348,16 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 		set_bit(I40E_VF_STAT_IWARPENA, &vf->vf_states);
 	}
 
-	if (pf->flags & I40E_FLAG_RSS_AQ_CAPABLE) {
-		if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ)
-			vfres->vf_offload_flags |=
-				I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ;
+	if (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PF) {
+		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RSS_PF;
 	} else {
-		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG;
+		if ((pf->flags & I40E_FLAG_RSS_AQ_CAPABLE) &&
+		    (vf->driver_caps & I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ))
+			vfres->vf_offload_flags |=
+					I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ;
+		else
+			vfres->vf_offload_flags |=
+					I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG;
 	}
 
 	if (pf->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE) {
@@ -1382,6 +1386,9 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 	vfres->num_vsis = num_vsis;
 	vfres->num_queue_pairs = vf->num_queue_pairs;
 	vfres->max_vectors = pf->hw.func_caps.num_msix_vectors_vf;
+	vfres->rss_key_size = I40E_HKEY_ARRAY_SIZE;
+	vfres->rss_lut_size = I40E_VF_HLUT_ARRAY_SIZE;
+
 	if (vf->lan_vsi_idx) {
 		vfres->vsi_res[0].vsi_id = vf->lan_vsi_id;
 		vfres->vsi_res[0].vsi_type = I40E_VSI_SRIOV;
@@ -2042,6 +2049,139 @@ error_param:
 }
 
 /**
+ * i40e_vc_config_rss_key
+ * @vf: pointer to the VF info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * Configure the VF's RSS key
+ **/
+static int i40e_vc_config_rss_key(struct i40e_vf *vf, u8 *msg, u16 msglen)
+{
+	struct i40e_virtchnl_rss_key *vrk =
+		(struct i40e_virtchnl_rss_key *)msg;
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_vsi *vsi = NULL;
+	u16 vsi_id = vrk->vsi_id;
+	i40e_status aq_ret = 0;
+
+	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+	    !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+	    !i40e_vc_isvalid_vsi_id(vf, vsi_id) ||
+	    (vrk->key_len != I40E_HKEY_ARRAY_SIZE)) {
+		aq_ret = I40E_ERR_PARAM;
+		goto err;
+	}
+
+	vsi = pf->vsi[vf->lan_vsi_idx];
+	aq_ret = i40e_config_rss(vsi, vrk->key, NULL, 0);
+err:
+	/* send the response to the VF */
+	return i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_CONFIG_RSS_KEY,
+				       aq_ret);
+}
+
+/**
+ * i40e_vc_config_rss_lut
+ * @vf: pointer to the VF info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * Configure the VF's RSS LUT
+ **/
+static int i40e_vc_config_rss_lut(struct i40e_vf *vf, u8 *msg, u16 msglen)
+{
+	struct i40e_virtchnl_rss_lut *vrl =
+		(struct i40e_virtchnl_rss_lut *)msg;
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_vsi *vsi = NULL;
+	u16 vsi_id = vrl->vsi_id;
+	i40e_status aq_ret = 0;
+
+	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+	    !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
+	    !i40e_vc_isvalid_vsi_id(vf, vsi_id) ||
+	    (vrl->lut_entries != I40E_VF_HLUT_ARRAY_SIZE)) {
+		aq_ret = I40E_ERR_PARAM;
+		goto err;
+	}
+
+	vsi = pf->vsi[vf->lan_vsi_idx];
+	aq_ret = i40e_config_rss(vsi, NULL, vrl->lut, I40E_VF_HLUT_ARRAY_SIZE);
+	/* send the response to the VF */
+err:
+	return i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_CONFIG_RSS_LUT,
+				       aq_ret);
+}
+
+/**
+ * i40e_vc_get_rss_hena
+ * @vf: pointer to the VF info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * Return the RSS HENA bits allowed by the hardware
+ **/
+static int i40e_vc_get_rss_hena(struct i40e_vf *vf, u8 *msg, u16 msglen)
+{
+	struct i40e_virtchnl_rss_hena *vrh = NULL;
+	struct i40e_pf *pf = vf->pf;
+	i40e_status aq_ret = 0;
+	int len = 0;
+
+	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+	    !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps)) {
+		aq_ret = I40E_ERR_PARAM;
+		goto err;
+	}
+	len = sizeof(struct i40e_virtchnl_rss_hena);
+
+	vrh = kzalloc(len, GFP_KERNEL);
+	if (!vrh) {
+		aq_ret = I40E_ERR_NO_MEMORY;
+		len = 0;
+		goto err;
+	}
+	vrh->hena = i40e_pf_get_default_rss_hena(pf);
+err:
+	/* send the response back to the VF */
+	aq_ret = i40e_vc_send_msg_to_vf(vf, I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS,
+					aq_ret, (u8 *)vrh, len);
+	return aq_ret;
+}
+
+/**
+ * i40e_vc_set_rss_hena
+ * @vf: pointer to the VF info
+ * @msg: pointer to the msg buffer
+ * @msglen: msg length
+ *
+ * Set the RSS HENA bits for the VF
+ **/
+static int i40e_vc_set_rss_hena(struct i40e_vf *vf, u8 *msg, u16 msglen)
+{
+	struct i40e_virtchnl_rss_hena *vrh =
+		(struct i40e_virtchnl_rss_hena *)msg;
+	struct i40e_pf *pf = vf->pf;
+	struct i40e_hw *hw = &pf->hw;
+	i40e_status aq_ret = 0;
+
+	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
+	    !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps)) {
+		aq_ret = I40E_ERR_PARAM;
+		goto err;
+	}
+	i40e_write_rx_ctl(hw, I40E_VFQF_HENA1(0, vf->vf_id), (u32)vrh->hena);
+	i40e_write_rx_ctl(hw, I40E_VFQF_HENA1(1, vf->vf_id),
+			  (u32)(vrh->hena >> 32));
+
+	/* send the response to the VF */
+err:
+	return i40e_vc_send_resp_to_vf(vf, I40E_VIRTCHNL_OP_SET_RSS_HENA,
+				       aq_ret);
+}
+
+/**
  * i40e_vc_validate_vf_msg
  * @vf: pointer to the VF info
  * @msg: pointer to the msg buffer
@@ -2162,6 +2302,36 @@ static int i40e_vc_validate_vf_msg(struct i40e_vf *vf, u32 v_opcode,
 				sizeof(struct i40e_virtchnl_iwarp_qv_info));
 		}
 		break;
+	case I40E_VIRTCHNL_OP_CONFIG_RSS_KEY:
+		valid_len = sizeof(struct i40e_virtchnl_rss_key);
+		if (msglen >= valid_len) {
+			struct i40e_virtchnl_rss_key *vrk =
+				(struct i40e_virtchnl_rss_key *)msg;
+			if (vrk->key_len != I40E_HKEY_ARRAY_SIZE) {
+				err_msg_format = true;
+				break;
+			}
+			valid_len += vrk->key_len - 1;
+		}
+		break;
+	case I40E_VIRTCHNL_OP_CONFIG_RSS_LUT:
+		valid_len = sizeof(struct i40e_virtchnl_rss_lut);
+		if (msglen >= valid_len) {
+			struct i40e_virtchnl_rss_lut *vrl =
+				(struct i40e_virtchnl_rss_lut *)msg;
+			if (vrl->lut_entries != I40E_VF_HLUT_ARRAY_SIZE) {
+				err_msg_format = true;
+				break;
+			}
+			valid_len += vrl->lut_entries - 1;
+		}
+		break;
+	case I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS:
+		valid_len = 0;
+		break;
+	case I40E_VIRTCHNL_OP_SET_RSS_HENA:
+		valid_len = sizeof(struct i40e_virtchnl_rss_hena);
+		break;
 	/* These are always errors coming from the VF. */
 	case I40E_VIRTCHNL_OP_EVENT:
 	case I40E_VIRTCHNL_OP_UNKNOWN:
@@ -2260,6 +2430,19 @@ int i40e_vc_process_vf_msg(struct i40e_pf *pf, u16 vf_id, u32 v_opcode,
 	case I40E_VIRTCHNL_OP_RELEASE_IWARP_IRQ_MAP:
 		ret = i40e_vc_iwarp_qvmap_msg(vf, msg, msglen, false);
 		break;
+	case I40E_VIRTCHNL_OP_CONFIG_RSS_KEY:
+		ret = i40e_vc_config_rss_key(vf, msg, msglen);
+		break;
+	case I40E_VIRTCHNL_OP_CONFIG_RSS_LUT:
+		ret = i40e_vc_config_rss_lut(vf, msg, msglen);
+		break;
+	case I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS:
+		ret = i40e_vc_get_rss_hena(vf, msg, msglen);
+		break;
+	case I40E_VIRTCHNL_OP_SET_RSS_HENA:
+		ret = i40e_vc_set_rss_hena(vf, msg, msglen);
+		break;
+
 	case I40E_VIRTCHNL_OP_UNKNOWN:
 	default:
 		dev_err(&pf->pdev->dev, "Unsupported opcode %d from VF %d\n",
-- 
2.5.5

[net-next 04/15] i40evf: Don't Panic

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Under some circumstances the driver remove function may be called before
the driver is fully initialized. So we can't assume that we know where
our towel is at, or that all of the data structures are initialized.

To ensure that we don't panic, check that the vsi_res pointer is valid
before dereferencing it. Then drink beer and eat peanuts.

Change-ID: If697b4db57348e39f9538793e16aa755e3e1af03
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index e657ecc..017c83b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -256,8 +256,10 @@ struct i40evf_adapter {
 	bool netdev_registered;
 	bool link_up;
 	enum i40e_virtchnl_ops current_op;
-#define CLIENT_ENABLED(_a) ((_a)->vf_res->vf_offload_flags & \
-			    I40E_VIRTCHNL_VF_OFFLOAD_IWARP)
+#define CLIENT_ENABLED(_a) ((_a)->vf_res ? \
+			    (_a)->vf_res->vf_offload_flags & \
+				I40E_VIRTCHNL_VF_OFFLOAD_IWARP : \
+			    0)
 #define RSS_AQ(_a) ((_a)->vf_res->vf_offload_flags & \
 		    I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ)
 #define VLAN_ALLOWED(_a) ((_a)->vf_res->vf_offload_flags & \
-- 
2.5.5

[net-next 05/15] i40e: Code cleanup in i40e_add_fdir_ethtool

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

A little bit of code cleanup in prep for more cloud filter work.

Change-ID: I0dc33ce0d4c207944336a07437640fef920c100c
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 8a83d45..8e56c43 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2506,7 +2506,6 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 
 	if (!vsi)
 		return -EINVAL;
-
 	pf = vsi->back;
 
 	if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED))
@@ -2564,15 +2563,18 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 	input->src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
 
 	if (ntohl(fsp->m_ext.data[1])) {
-		if (ntohl(fsp->h_ext.data[1]) >= pf->num_alloc_vfs) {
-			netif_info(pf, drv, vsi->netdev, "Invalid VF id\n");
+		vf_id = ntohl(fsp->h_ext.data[1]);
+		if (vf_id >= pf->num_alloc_vfs) {
+			netif_info(pf, drv, vsi->netdev,
+				   "Invalid VF id %d\n", vf_id);
 			goto free_input;
 		}
-		vf_id = ntohl(fsp->h_ext.data[1]);
 		/* Find vsi id from vf id and override dest vsi */
 		input->dest_vsi = pf->vf[vf_id].lan_vsi_id;
 		if (input->q_index >= pf->vf[vf_id].num_queue_pairs) {
-			netif_info(pf, drv, vsi->netdev, "Invalid queue id\n");
+			netif_info(pf, drv, vsi->netdev,
+				   "Invalid queue id %d for VF %d\n",
+				   input->q_index, vf_id);
 			goto free_input;
 		}
 	}
-- 
2.5.5

[net-next 06/15] i40e: Specify AQ event opcode to wait for

[Jeff Kirsher]

From: Shannon Nelson <shannon.nelson@intel.com>

To add a little flexibility to the nvmupdate facility, this code adds the
ability to specify an AQ event opcode to wait on after the Exec_AQ request.

Change-ID: Iddbfd63c3de8df3edb9d3e90678b08989bc4946e
Signed-off-by: Shannon Nelson <shannon.nelson@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_nvm.c    | 49 +++++++++++++++++++++++----
 drivers/net/ethernet/intel/i40e/i40e_type.h   |  1 +
 drivers/net/ethernet/intel/i40evf/i40e_type.h |  1 +
 3 files changed, 44 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_nvm.c b/drivers/net/ethernet/intel/i40e/i40e_nvm.c
index f2cea3d..954efe3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_nvm.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_nvm.c
@@ -693,10 +693,10 @@ i40e_status i40e_nvmupd_command(struct i40e_hw *hw,
 	/* early check for status command and debug msgs */
 	upd_cmd = i40e_nvmupd_validate_command(hw, cmd, perrno);
 
-	i40e_debug(hw, I40E_DEBUG_NVM, "%s state %d nvm_release_on_hold %d cmd 0x%08x config 0x%08x offset 0x%08x data_size 0x%08x\n",
+	i40e_debug(hw, I40E_DEBUG_NVM, "%s state %d nvm_release_on_hold %d opc 0x%04x cmd 0x%08x config 0x%08x offset 0x%08x data_size 0x%08x\n",
 		   i40e_nvm_update_state_str[upd_cmd],
 		   hw->nvmupd_state,
-		   hw->nvm_release_on_done,
+		   hw->nvm_release_on_done, hw->nvm_wait_opcode,
 		   cmd->command, cmd->config, cmd->offset, cmd->data_size);
 
 	if (upd_cmd == I40E_NVMUPD_INVALID) {
@@ -710,7 +710,18 @@ i40e_status i40e_nvmupd_command(struct i40e_hw *hw,
 	 * going into the state machine
 	 */
 	if (upd_cmd == I40E_NVMUPD_STATUS) {
+		if (!cmd->data_size) {
+			*perrno = -EFAULT;
+			return I40E_ERR_BUF_TOO_SHORT;
+		}
+
 		bytes[0] = hw->nvmupd_state;
+
+		if (cmd->data_size >= 4) {
+			bytes[1] = 0;
+			*((u16 *)&bytes[2]) = hw->nvm_wait_opcode;
+		}
+
 		return 0;
 	}
 
@@ -729,6 +740,14 @@ i40e_status i40e_nvmupd_command(struct i40e_hw *hw,
 
 	case I40E_NVMUPD_STATE_INIT_WAIT:
 	case I40E_NVMUPD_STATE_WRITE_WAIT:
+		/* if we need to stop waiting for an event, clear
+		 * the wait info and return before doing anything else
+		 */
+		if (cmd->offset == 0xffff) {
+			i40e_nvmupd_check_wait_event(hw, hw->nvm_wait_opcode);
+			return 0;
+		}
+
 		status = I40E_ERR_NOT_READY;
 		*perrno = -EBUSY;
 		break;
@@ -800,6 +819,7 @@ static i40e_status i40e_nvmupd_state_init(struct i40e_hw *hw,
 				i40e_release_nvm(hw);
 			} else {
 				hw->nvm_release_on_done = true;
+				hw->nvm_wait_opcode = i40e_aqc_opc_nvm_erase;
 				hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
 			}
 		}
@@ -816,6 +836,7 @@ static i40e_status i40e_nvmupd_state_init(struct i40e_hw *hw,
 				i40e_release_nvm(hw);
 			} else {
 				hw->nvm_release_on_done = true;
+				hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 				hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
 			}
 		}
@@ -828,10 +849,12 @@ static i40e_status i40e_nvmupd_state_init(struct i40e_hw *hw,
 						     hw->aq.asq_last_status);
 		} else {
 			status = i40e_nvmupd_nvm_write(hw, cmd, bytes, perrno);
-			if (status)
+			if (status) {
 				i40e_release_nvm(hw);
-			else
+			} else {
+				hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 				hw->nvmupd_state = I40E_NVMUPD_STATE_WRITE_WAIT;
+			}
 		}
 		break;
 
@@ -850,6 +873,7 @@ static i40e_status i40e_nvmupd_state_init(struct i40e_hw *hw,
 				i40e_release_nvm(hw);
 			} else {
 				hw->nvm_release_on_done = true;
+				hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 				hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
 			}
 		}
@@ -940,8 +964,10 @@ retry:
 	switch (upd_cmd) {
 	case I40E_NVMUPD_WRITE_CON:
 		status = i40e_nvmupd_nvm_write(hw, cmd, bytes, perrno);
-		if (!status)
+		if (!status) {
+			hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 			hw->nvmupd_state = I40E_NVMUPD_STATE_WRITE_WAIT;
+		}
 		break;
 
 	case I40E_NVMUPD_WRITE_LCB:
@@ -954,6 +980,7 @@ retry:
 			hw->nvmupd_state = I40E_NVMUPD_STATE_INIT;
 		} else {
 			hw->nvm_release_on_done = true;
+			hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 			hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
 		}
 		break;
@@ -967,6 +994,7 @@ retry:
 				   -EIO;
 			hw->nvmupd_state = I40E_NVMUPD_STATE_INIT;
 		} else {
+			hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 			hw->nvmupd_state = I40E_NVMUPD_STATE_WRITE_WAIT;
 		}
 		break;
@@ -981,6 +1009,7 @@ retry:
 			hw->nvmupd_state = I40E_NVMUPD_STATE_INIT;
 		} else {
 			hw->nvm_release_on_done = true;
+			hw->nvm_wait_opcode = i40e_aqc_opc_nvm_update;
 			hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
 		}
 		break;
@@ -1036,14 +1065,14 @@ retry:
  **/
 void i40e_nvmupd_check_wait_event(struct i40e_hw *hw, u16 opcode)
 {
-	if (opcode == i40e_aqc_opc_nvm_erase ||
-	    opcode == i40e_aqc_opc_nvm_update) {
+	if (opcode == hw->nvm_wait_opcode) {
 		i40e_debug(hw, I40E_DEBUG_NVM,
 			   "NVMUPD: clearing wait on opcode 0x%04x\n", opcode);
 		if (hw->nvm_release_on_done) {
 			i40e_release_nvm(hw);
 			hw->nvm_release_on_done = false;
 		}
+		hw->nvm_wait_opcode = 0;
 
 		switch (hw->nvmupd_state) {
 		case I40E_NVMUPD_STATE_INIT_WAIT:
@@ -1220,6 +1249,12 @@ static i40e_status i40e_nvmupd_exec_aq(struct i40e_hw *hw,
 		*perrno = i40e_aq_rc_to_posix(status, hw->aq.asq_last_status);
 	}
 
+	/* should we wait for a followup event? */
+	if (cmd->offset) {
+		hw->nvm_wait_opcode = cmd->offset;
+		hw->nvmupd_state = I40E_NVMUPD_STATE_INIT_WAIT;
+	}
+
 	return status;
 }
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index 793036b..bb57cd9 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -550,6 +550,7 @@ struct i40e_hw {
 	struct i40e_aq_desc nvm_wb_desc;
 	struct i40e_virt_mem nvm_buff;
 	bool nvm_release_on_done;
+	u16 nvm_wait_opcode;
 
 	/* HMC info */
 	struct i40e_hmc_info hmc; /* HMC info struct */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_type.h b/drivers/net/ethernet/intel/i40evf/i40e_type.h
index 4a78c18..b720713 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_type.h
@@ -523,6 +523,7 @@ struct i40e_hw {
 	struct i40e_aq_desc nvm_wb_desc;
 	struct i40e_virt_mem nvm_buff;
 	bool nvm_release_on_done;
+	u16 nvm_wait_opcode;
 
 	/* HMC info */
 	struct i40e_hmc_info hmc; /* HMC info struct */
-- 
2.5.5

[net-next 07/15] i40evf: Allow PF driver to configure RSS

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

If the PF driver reports proper support, allow the PF driver to
configure RSS on the behalf of the VF driver. This will allow for RSS
support on future hardware without changes to the VF driver.

Unfortunately, the old RSS code still needs to stay as the driver needs
to be compatible with PF drivers that don't support this interface. But
this change still simplifies the data structures a bunch and makes this
code simpler to read and maintain.

Change-ID: I0375aad40788ecdc0cb24d5cfeccf07804e69771
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf.h         |  30 +-
 drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c | 121 ++++----
 drivers/net/ethernet/intel/i40evf/i40evf_main.c    | 308 +++++++--------------
 .../net/ethernet/intel/i40evf/i40evf_virtchnl.c    | 119 ++++++++
 4 files changed, 294 insertions(+), 284 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index 017c83b..63f7aae 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -67,8 +67,6 @@ struct i40e_vsi {
 	u16 rx_itr_setting;
 	u16 tx_itr_setting;
 	u16 qs_handle;
-	u8 *rss_hkey_user; /* User configured hash keys */
-	u8 *rss_lut_user;  /* User configured lookup table entries */
 };
 
 /* How many Rx Buffers do we bundle into one write to the hardware ? */
@@ -239,8 +237,13 @@ struct i40evf_adapter {
 #define I40EVF_FLAG_AQ_CONFIGURE_QUEUES		BIT(6)
 #define I40EVF_FLAG_AQ_MAP_VECTORS		BIT(7)
 #define I40EVF_FLAG_AQ_HANDLE_RESET		BIT(8)
-#define I40EVF_FLAG_AQ_CONFIGURE_RSS		BIT(9)
+#define I40EVF_FLAG_AQ_CONFIGURE_RSS		BIT(9)	/* direct AQ config */
 #define I40EVF_FLAG_AQ_GET_CONFIG		BIT(10)
+/* Newer style, RSS done by the PF so we can ignore hardware vagaries. */
+#define I40EVF_FLAG_AQ_GET_HENA			BIT(11)
+#define I40EVF_FLAG_AQ_SET_HENA			BIT(12)
+#define I40EVF_FLAG_AQ_SET_RSS_KEY		BIT(13)
+#define I40EVF_FLAG_AQ_SET_RSS_LUT		BIT(14)
 
 	/* OS defined structs */
 	struct net_device *netdev;
@@ -260,8 +263,14 @@ struct i40evf_adapter {
 			    (_a)->vf_res->vf_offload_flags & \
 				I40E_VIRTCHNL_VF_OFFLOAD_IWARP : \
 			    0)
+/* RSS by the PF should be preferred over RSS via other methods. */
+#define RSS_PF(_a) ((_a)->vf_res->vf_offload_flags & \
+		    I40E_VIRTCHNL_VF_OFFLOAD_RSS_PF)
 #define RSS_AQ(_a) ((_a)->vf_res->vf_offload_flags & \
 		    I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ)
+#define RSS_REG(_a) (!((_a)->vf_res->vf_offload_flags & \
+		       (I40E_VIRTCHNL_VF_OFFLOAD_RSS_AQ | \
+			I40E_VIRTCHNL_VF_OFFLOAD_RSS_PF)))
 #define VLAN_ALLOWED(_a) ((_a)->vf_res->vf_offload_flags & \
 			  I40E_VIRTCHNL_VF_OFFLOAD_VLAN)
 	struct i40e_virtchnl_vf_resource *vf_res; /* incl. all VSIs */
@@ -273,6 +282,12 @@ struct i40evf_adapter {
 	struct i40e_eth_stats current_stats;
 	struct i40e_vsi vsi;
 	u32 aq_wait_count;
+	/* RSS stuff */
+	u64 hena;
+	u16 rss_key_size;
+	u16 rss_lut_size;
+	u8 *rss_key;
+	u8 *rss_lut;
 };
 
 
@@ -316,11 +331,12 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter);
 void i40evf_set_promiscuous(struct i40evf_adapter *adapter, int flags);
 void i40evf_request_stats(struct i40evf_adapter *adapter);
 void i40evf_request_reset(struct i40evf_adapter *adapter);
+void i40evf_get_hena(struct i40evf_adapter *adapter);
+void i40evf_set_hena(struct i40evf_adapter *adapter);
+void i40evf_set_rss_key(struct i40evf_adapter *adapter);
+void i40evf_set_rss_lut(struct i40evf_adapter *adapter);
 void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 				enum i40e_virtchnl_ops v_opcode,
 				i40e_status v_retval, u8 *msg, u16 msglen);
-int i40evf_config_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut,
-		      u16 lut_size);
-int i40evf_get_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut,
-		   u16 lut_size);
+int i40evf_config_rss(struct i40evf_adapter *adapter);
 #endif /* _I40EVF_H_ */
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
index dd4430a..9f7657c 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_ethtool.c
@@ -387,20 +387,16 @@ static int i40evf_set_coalesce(struct net_device *netdev,
 static int i40evf_get_rss_hash_opts(struct i40evf_adapter *adapter,
 				    struct ethtool_rxnfc *cmd)
 {
-	struct i40e_hw *hw = &adapter->hw;
-	u64 hena = (u64)rd32(hw, I40E_VFQF_HENA(0)) |
-		   ((u64)rd32(hw, I40E_VFQF_HENA(1)) << 32);
-
 	/* We always hash on IP src and dest addresses */
 	cmd->data = RXH_IP_SRC | RXH_IP_DST;
 
 	switch (cmd->flow_type) {
 	case TCP_V4_FLOW:
-		if (hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP))
+		if (adapter->hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP))
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	case UDP_V4_FLOW:
-		if (hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP))
+		if (adapter->hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP))
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 
@@ -412,11 +408,11 @@ static int i40evf_get_rss_hash_opts(struct i40evf_adapter *adapter,
 		break;
 
 	case TCP_V6_FLOW:
-		if (hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP))
+		if (adapter->hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP))
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 	case UDP_V6_FLOW:
-		if (hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP))
+		if (adapter->hena & BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP))
 			cmd->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
 		break;
 
@@ -476,9 +472,6 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 	struct i40e_hw *hw = &adapter->hw;
 	u32 flags = adapter->vf_res->vf_offload_flags;
 
-	u64 hena = (u64)rd32(hw, I40E_VFQF_HENA(0)) |
-		   ((u64)rd32(hw, I40E_VFQF_HENA(1)) << 32);
-
 	/* RSS does not support anything other than hashing
 	 * to queues on src and dst IPs and ports
 	 */
@@ -495,10 +488,11 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 	case TCP_V4_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
-				hena |=
+				adapter->hena |=
 			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK);
 
-			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
+			adapter->hena |=
+		 BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_TCP);
 		} else {
 			return -EINVAL;
 		}
@@ -506,10 +500,11 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 	case TCP_V6_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
-				hena |=
+				adapter->hena |=
 			   BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK);
 
-			hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
+			adapter->hena |=
+		 BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_TCP);
 		} else {
 			return -EINVAL;
 		}
@@ -517,11 +512,12 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 	case UDP_V4_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
-				hena |=
+				adapter->hena |=
 			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) |
 			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP);
 
-			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
+			adapter->hena |=
+				(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
 		} else {
 			return -EINVAL;
@@ -530,11 +526,12 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 	case UDP_V6_FLOW:
 		if (nfc->data & (RXH_L4_B_0_1 | RXH_L4_B_2_3)) {
 			if (flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
-				hena |=
+				adapter->hena |=
 			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) |
 			    BIT_ULL(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP);
 
-			hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
+			adapter->hena |=
+				(BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_UDP) |
 				 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
 		} else {
 			return -EINVAL;
@@ -547,7 +544,7 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 		if ((nfc->data & RXH_L4_B_0_1) ||
 		    (nfc->data & RXH_L4_B_2_3))
 			return -EINVAL;
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER);
+		adapter->hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER);
 		break;
 	case AH_ESP_V6_FLOW:
 	case AH_V6_FLOW:
@@ -556,23 +553,27 @@ static int i40evf_set_rss_hash_opt(struct i40evf_adapter *adapter,
 		if ((nfc->data & RXH_L4_B_0_1) ||
 		    (nfc->data & RXH_L4_B_2_3))
 			return -EINVAL;
-		hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER);
+		adapter->hena |= BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER);
 		break;
 	case IPV4_FLOW:
-		hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
-			 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
+		adapter->hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV4_OTHER) |
+				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV4));
 		break;
 	case IPV6_FLOW:
-		hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
-			 BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
+		adapter->hena |= (BIT_ULL(I40E_FILTER_PCTYPE_NONF_IPV6_OTHER) |
+				  BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6));
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
-	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
-	i40e_flush(hw);
+	if (RSS_PF(adapter)) {
+		adapter->aq_required = I40EVF_FLAG_AQ_SET_HENA;
+	} else {
+		wr32(hw, I40E_VFQF_HENA(0), (u32)adapter->hena);
+		wr32(hw, I40E_VFQF_HENA(1), (u32)(adapter->hena >> 32));
+		i40e_flush(hw);
+	}
 
 	return 0;
 }
@@ -624,6 +625,19 @@ static void i40evf_get_channels(struct net_device *netdev,
 }
 
 /**
+ * i40evf_get_rxfh_key_size - get the RSS hash key size
+ * @netdev: network interface device structure
+ *
+ * Returns the table size.
+ **/
+static u32 i40evf_get_rxfh_key_size(struct net_device *netdev)
+{
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+
+	return adapter->rss_key_size;
+}
+
+/**
  * i40evf_get_rxfh_indir_size - get the rx flow hash indirection table size
  * @netdev: network interface device structure
  *
@@ -631,7 +645,9 @@ static void i40evf_get_channels(struct net_device *netdev,
  **/
 static u32 i40evf_get_rxfh_indir_size(struct net_device *netdev)
 {
-	return (I40E_VFQF_HLUT_MAX_INDEX + 1) * 4;
+	struct i40evf_adapter *adapter = netdev_priv(netdev);
+
+	return adapter->rss_lut_size;
 }
 
 /**
@@ -646,9 +662,6 @@ static int i40evf_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 			   u8 *hfunc)
 {
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
-	struct i40e_vsi *vsi = &adapter->vsi;
-	u8 *seed = NULL, *lut;
-	int ret;
 	u16 i;
 
 	if (hfunc)
@@ -656,24 +669,13 @@ static int i40evf_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 	if (!indir)
 		return 0;
 
-	seed = key;
-
-	lut = kzalloc(I40EVF_HLUT_ARRAY_SIZE, GFP_KERNEL);
-	if (!lut)
-		return -ENOMEM;
-
-	ret = i40evf_get_rss(vsi, seed, lut, I40EVF_HLUT_ARRAY_SIZE);
-	if (ret)
-		goto out;
+	memcpy(key, adapter->rss_key, adapter->rss_key_size);
 
 	/* Each 32 bits pointed by 'indir' is stored with a lut entry */
-	for (i = 0; i < I40EVF_HLUT_ARRAY_SIZE; i++)
-		indir[i] = (u32)lut[i];
+	for (i = 0; i < adapter->rss_lut_size; i++)
+		indir[i] = (u32)adapter->rss_lut[i];
 
-out:
-	kfree(lut);
-
-	return ret;
+	return 0;
 }
 
 /**
@@ -689,8 +691,6 @@ static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir,
 			   const u8 *key, const u8 hfunc)
 {
 	struct i40evf_adapter *adapter = netdev_priv(netdev);
-	struct i40e_vsi *vsi = &adapter->vsi;
-	u8 *seed = NULL;
 	u16 i;
 
 	/* We do not allow change in unsupported parameters */
@@ -701,28 +701,14 @@ static int i40evf_set_rxfh(struct net_device *netdev, const u32 *indir,
 		return 0;
 
 	if (key) {
-		if (!vsi->rss_hkey_user) {
-			vsi->rss_hkey_user = kzalloc(I40EVF_HKEY_ARRAY_SIZE,
-						     GFP_KERNEL);
-			if (!vsi->rss_hkey_user)
-				return -ENOMEM;
-		}
-		memcpy(vsi->rss_hkey_user, key, I40EVF_HKEY_ARRAY_SIZE);
-		seed = vsi->rss_hkey_user;
-	}
-	if (!vsi->rss_lut_user) {
-		vsi->rss_lut_user = kzalloc(I40EVF_HLUT_ARRAY_SIZE,
-					    GFP_KERNEL);
-		if (!vsi->rss_lut_user)
-			return -ENOMEM;
+		memcpy(adapter->rss_key, key, adapter->rss_key_size);
 	}
 
 	/* Each 32 bits pointed by 'indir' is stored with a lut entry */
-	for (i = 0; i < I40EVF_HLUT_ARRAY_SIZE; i++)
-		vsi->rss_lut_user[i] = (u8)(indir[i]);
+	for (i = 0; i < adapter->rss_lut_size; i++)
+		adapter->rss_lut[i] = (u8)(indir[i]);
 
-	return i40evf_config_rss(vsi, seed, vsi->rss_lut_user,
-				 I40EVF_HLUT_ARRAY_SIZE);
+	return i40evf_config_rss(adapter);
 }
 
 /**
@@ -794,6 +780,7 @@ static const struct ethtool_ops i40evf_ethtool_ops = {
 	.get_rxfh		= i40evf_get_rxfh,
 	.set_rxfh		= i40evf_set_rxfh,
 	.get_channels		= i40evf_get_channels,
+	.get_rxfh_key_size	= i40evf_get_rxfh_key_size,
 };
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 806da26..af53159 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1224,24 +1224,18 @@ out:
 }
 
 /**
- * i40e_config_rss_aq - Prepare for RSS using AQ commands
- * @vsi: vsi structure
- * @seed: RSS hash seed
- * @lut: Lookup table
- * @lut_size: Lookup table size
+ * i40e_config_rss_aq - Configure RSS keys and lut by using AQ commands
+ * @adapter: board private structure
  *
  * Return 0 on success, negative on failure
  **/
-static int i40evf_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed,
-				u8 *lut, u16 lut_size)
+static int i40evf_config_rss_aq(struct i40evf_adapter *adapter)
 {
-	struct i40evf_adapter *adapter = vsi->back;
+	struct i40e_aqc_get_set_rss_key_data *rss_key =
+		(struct i40e_aqc_get_set_rss_key_data *)adapter->rss_key;
 	struct i40e_hw *hw = &adapter->hw;
 	int ret = 0;
 
-	if (!vsi->id)
-		return -EINVAL;
-
 	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
 		/* bail because we already have a command pending */
 		dev_err(&adapter->pdev->dev, "Cannot configure RSS, command %d pending\n",
@@ -1249,198 +1243,82 @@ static int i40evf_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed,
 		return -EBUSY;
 	}
 
-	if (seed) {
-		struct i40e_aqc_get_set_rss_key_data *rss_key =
-			(struct i40e_aqc_get_set_rss_key_data *)seed;
-		ret = i40evf_aq_set_rss_key(hw, vsi->id, rss_key);
-		if (ret) {
-			dev_err(&adapter->pdev->dev, "Cannot set RSS key, err %s aq_err %s\n",
-				i40evf_stat_str(hw, ret),
-				i40evf_aq_str(hw, hw->aq.asq_last_status));
-			return ret;
-		}
+	ret = i40evf_aq_set_rss_key(hw, adapter->vsi.id, rss_key);
+	if (ret) {
+		dev_err(&adapter->pdev->dev, "Cannot set RSS key, err %s aq_err %s\n",
+			i40evf_stat_str(hw, ret),
+			i40evf_aq_str(hw, hw->aq.asq_last_status));
+		return ret;
+
 	}
 
-	if (lut) {
-		ret = i40evf_aq_set_rss_lut(hw, vsi->id, false, lut, lut_size);
-		if (ret) {
-			dev_err(&adapter->pdev->dev,
-				"Cannot set RSS lut, err %s aq_err %s\n",
-				i40evf_stat_str(hw, ret),
-				i40evf_aq_str(hw, hw->aq.asq_last_status));
-			return ret;
-		}
+	ret = i40evf_aq_set_rss_lut(hw, adapter->vsi.id, false,
+				    adapter->rss_lut, adapter->rss_lut_size);
+	if (ret) {
+		dev_err(&adapter->pdev->dev, "Cannot set RSS lut, err %s aq_err %s\n",
+			i40evf_stat_str(hw, ret),
+			i40evf_aq_str(hw, hw->aq.asq_last_status));
 	}
 
 	return ret;
+
 }
 
 /**
  * i40evf_config_rss_reg - Configure RSS keys and lut by writing registers
- * @vsi: Pointer to vsi structure
- * @seed: RSS hash seed
- * @lut: Lookup table
- * @lut_size: Lookup table size
+ * @adapter: board private structure
  *
  * Returns 0 on success, negative on failure
  **/
-static int i40evf_config_rss_reg(struct i40e_vsi *vsi, const u8 *seed,
-				 const u8 *lut, u16 lut_size)
+static int i40evf_config_rss_reg(struct i40evf_adapter *adapter)
 {
-	struct i40evf_adapter *adapter = vsi->back;
 	struct i40e_hw *hw = &adapter->hw;
+	u32 *dw;
 	u16 i;
 
-	if (seed) {
-		u32 *seed_dw = (u32 *)seed;
-
-		for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++)
-			wr32(hw, I40E_VFQF_HKEY(i), seed_dw[i]);
-	}
-
-	if (lut) {
-		u32 *lut_dw = (u32 *)lut;
+	dw = (u32 *)adapter->rss_key;
+	for (i = 0; i <= adapter->rss_key_size / 4; i++)
+		wr32(hw, I40E_VFQF_HKEY(i), dw[i]);
 
-		if (lut_size != I40EVF_HLUT_ARRAY_SIZE)
-			return -EINVAL;
+	dw = (u32 *)adapter->rss_lut;
+	for (i = 0; i <= adapter->rss_lut_size / 4; i++)
+		wr32(hw, I40E_VFQF_HLUT(i), dw[i]);
 
-		for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++)
-			wr32(hw, I40E_VFQF_HLUT(i), lut_dw[i]);
-	}
 	i40e_flush(hw);
 
 	return 0;
 }
 
 /**
- *  * i40evf_get_rss_aq - Get RSS keys and lut by using AQ commands
- *  @vsi: Pointer to vsi structure
- *  @seed: RSS hash seed
- *  @lut: Lookup table
- *  @lut_size: Lookup table size
- *
- *  Return 0 on success, negative on failure
- **/
-static int i40evf_get_rss_aq(struct i40e_vsi *vsi, const u8 *seed,
-			     u8 *lut, u16 lut_size)
-{
-	struct i40evf_adapter *adapter = vsi->back;
-	struct i40e_hw *hw = &adapter->hw;
-	int ret = 0;
-
-	if (seed) {
-		ret = i40evf_aq_get_rss_key(hw, vsi->id,
-			(struct i40e_aqc_get_set_rss_key_data *)seed);
-		if (ret) {
-			dev_err(&adapter->pdev->dev,
-				"Cannot get RSS key, err %s aq_err %s\n",
-				i40evf_stat_str(hw, ret),
-				i40evf_aq_str(hw, hw->aq.asq_last_status));
-			return ret;
-		}
-	}
-
-	if (lut) {
-		ret = i40evf_aq_get_rss_lut(hw, vsi->id, false, lut, lut_size);
-		if (ret) {
-			dev_err(&adapter->pdev->dev,
-				"Cannot get RSS lut, err %s aq_err %s\n",
-				i40evf_stat_str(hw, ret),
-				i40evf_aq_str(hw, hw->aq.asq_last_status));
-			return ret;
-		}
-	}
-
-	return ret;
-}
-
-/**
- *  * i40evf_get_rss_reg - Get RSS keys and lut by reading registers
- *  @vsi: Pointer to vsi structure
- *  @seed: RSS hash seed
- *  @lut: Lookup table
- *  @lut_size: Lookup table size
- *
- *  Returns 0 on success, negative on failure
- **/
-static int i40evf_get_rss_reg(struct i40e_vsi *vsi, const u8 *seed,
-			      const u8 *lut, u16 lut_size)
-{
-	struct i40evf_adapter *adapter = vsi->back;
-	struct i40e_hw *hw = &adapter->hw;
-	u16 i;
-
-	if (seed) {
-		u32 *seed_dw = (u32 *)seed;
-
-		for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++)
-			seed_dw[i] = rd32(hw, I40E_VFQF_HKEY(i));
-	}
-
-	if (lut) {
-		u32 *lut_dw = (u32 *)lut;
-
-		if (lut_size != I40EVF_HLUT_ARRAY_SIZE)
-			return -EINVAL;
-
-		for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++)
-			lut_dw[i] = rd32(hw, I40E_VFQF_HLUT(i));
-	}
-
-	return 0;
-}
-
-/**
  * i40evf_config_rss - Configure RSS keys and lut
- * @vsi: Pointer to vsi structure
- * @seed: RSS hash seed
- * @lut: Lookup table
- * @lut_size: Lookup table size
- *
- * Returns 0 on success, negative on failure
- **/
-int i40evf_config_rss(struct i40e_vsi *vsi, const u8 *seed,
-		      u8 *lut, u16 lut_size)
-{
-	struct i40evf_adapter *adapter = vsi->back;
-
-	if (RSS_AQ(adapter))
-		return i40evf_config_rss_aq(vsi, seed, lut, lut_size);
-	else
-		return i40evf_config_rss_reg(vsi, seed, lut, lut_size);
-}
-
-/**
- * i40evf_get_rss - Get RSS keys and lut
- * @vsi: Pointer to vsi structure
- * @seed: RSS hash seed
- * @lut: Lookup table
- * @lut_size: Lookup table size
+ * @adapter: board private structure
  *
  * Returns 0 on success, negative on failure
  **/
-int i40evf_get_rss(struct i40e_vsi *vsi, const u8 *seed, u8 *lut, u16 lut_size)
+int i40evf_config_rss(struct i40evf_adapter *adapter)
 {
-	struct i40evf_adapter *adapter = vsi->back;
 
-	if (RSS_AQ(adapter))
-		return i40evf_get_rss_aq(vsi, seed, lut, lut_size);
-	else
-		return i40evf_get_rss_reg(vsi, seed, lut, lut_size);
+	if (RSS_PF(adapter)) {
+		adapter->aq_required |= I40EVF_FLAG_AQ_SET_RSS_LUT |
+					I40EVF_FLAG_AQ_SET_RSS_KEY;
+		return 0;
+	} else if (RSS_AQ(adapter)) {
+		return i40evf_config_rss_aq(adapter);
+	} else {
+		return i40evf_config_rss_reg(adapter);
+	}
 }
 
 /**
  * i40evf_fill_rss_lut - Fill the lut with default values
- * @lut: Lookup table to be filled with
- * @rss_table_size: Lookup table size
- * @rss_size: Range of queue number for hashing
+ * @adapter: board private structure
  **/
-static void i40evf_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size)
+static void i40evf_fill_rss_lut(struct i40evf_adapter *adapter)
 {
 	u16 i;
 
-	for (i = 0; i < rss_table_size; i++)
-		lut[i] = i % rss_size;
+	for (i = 0; i < adapter->rss_lut_size; i++)
+		adapter->rss_lut[i] = i % adapter->num_active_queues;
 }
 
 /**
@@ -1451,42 +1329,25 @@ static void i40evf_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size)
  **/
 static int i40evf_init_rss(struct i40evf_adapter *adapter)
 {
-	struct i40e_vsi *vsi = &adapter->vsi;
 	struct i40e_hw *hw = &adapter->hw;
-	u8 seed[I40EVF_HKEY_ARRAY_SIZE];
-	u64 hena;
-	u8 *lut;
 	int ret;
 
-	/* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */
-	if (adapter->vf_res->vf_offload_flags &
-					I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
-		hena = I40E_DEFAULT_RSS_HENA_EXPANDED;
-	else
-		hena = I40E_DEFAULT_RSS_HENA;
-	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
-	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
+	if (!RSS_PF(adapter)) {
+		/* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */
+		if (adapter->vf_res->vf_offload_flags &
+		    I40E_VIRTCHNL_VF_OFFLOAD_RSS_PCTYPE_V2)
+			adapter->hena = I40E_DEFAULT_RSS_HENA_EXPANDED;
+		else
+			adapter->hena = I40E_DEFAULT_RSS_HENA;
 
-	lut = kzalloc(I40EVF_HLUT_ARRAY_SIZE, GFP_KERNEL);
-	if (!lut)
-		return -ENOMEM;
+		wr32(hw, I40E_VFQF_HENA(0), (u32)adapter->hena);
+		wr32(hw, I40E_VFQF_HENA(1), (u32)(adapter->hena >> 32));
+	}
 
-	/* Use user configured lut if there is one, otherwise use default */
-	if (vsi->rss_lut_user)
-		memcpy(lut, vsi->rss_lut_user, I40EVF_HLUT_ARRAY_SIZE);
-	else
-		i40evf_fill_rss_lut(lut, I40EVF_HLUT_ARRAY_SIZE,
-				    adapter->num_active_queues);
+	i40evf_fill_rss_lut(adapter);
 
-	/* Use user configured hash key if there is one, otherwise
-	 * user default.
-	 */
-	if (vsi->rss_hkey_user)
-		memcpy(seed, vsi->rss_hkey_user, I40EVF_HKEY_ARRAY_SIZE);
-	else
-		netdev_rss_key_fill((void *)seed, I40EVF_HKEY_ARRAY_SIZE);
-	ret = i40evf_config_rss(vsi, seed, lut, I40EVF_HLUT_ARRAY_SIZE);
-	kfree(lut);
+	netdev_rss_key_fill((void *)adapter->rss_key, adapter->rss_key_size);
+	ret = i40evf_config_rss(adapter);
 
 	return ret;
 }
@@ -1601,19 +1462,16 @@ err_set_interrupt:
 }
 
 /**
- * i40evf_clear_rss_config_user - Clear user configurations of RSS
- * @vsi: Pointer to VSI structure
+ * i40evf_free_rss - Free memory used by RSS structs
+ * @adapter: board private structure
  **/
-static void i40evf_clear_rss_config_user(struct i40e_vsi *vsi)
+static void i40evf_free_rss(struct i40evf_adapter *adapter)
 {
-	if (!vsi)
-		return;
-
-	kfree(vsi->rss_hkey_user);
-	vsi->rss_hkey_user = NULL;
+	kfree(adapter->rss_key);
+	adapter->rss_key = NULL;
 
-	kfree(vsi->rss_lut_user);
-	vsi->rss_lut_user = NULL;
+	kfree(adapter->rss_lut);
+	adapter->rss_lut = NULL;
 }
 
 /**
@@ -1747,6 +1605,22 @@ static void i40evf_watchdog_task(struct work_struct *work)
 		adapter->aq_required &= ~I40EVF_FLAG_AQ_CONFIGURE_RSS;
 		goto watchdog_done;
 	}
+	if (adapter->aq_required & I40EVF_FLAG_AQ_GET_HENA) {
+		i40evf_get_hena(adapter);
+		goto watchdog_done;
+	}
+	if (adapter->aq_required & I40EVF_FLAG_AQ_SET_HENA) {
+		i40evf_set_hena(adapter);
+		goto watchdog_done;
+	}
+	if (adapter->aq_required & I40EVF_FLAG_AQ_SET_RSS_KEY) {
+		i40evf_set_rss_key(adapter);
+		goto watchdog_done;
+	}
+	if (adapter->aq_required & I40EVF_FLAG_AQ_SET_RSS_LUT) {
+		i40evf_set_rss_lut(adapter);
+		goto watchdog_done;
+	}
 
 	if (adapter->state == __I40EVF_RUNNING)
 		i40evf_request_stats(adapter);
@@ -2325,6 +2199,7 @@ int i40evf_process_config(struct i40evf_adapter *adapter)
 {
 	struct i40e_virtchnl_vf_resource *vfres = adapter->vf_res;
 	struct net_device *netdev = adapter->netdev;
+	struct i40e_vsi *vsi = &adapter->vsi;
 	int i;
 
 	/* got VF config message back from PF, now we can parse it */
@@ -2381,8 +2256,16 @@ int i40evf_process_config(struct i40evf_adapter *adapter)
 				       ITR_REG_TO_USEC(I40E_ITR_RX_DEF));
 	adapter->vsi.tx_itr_setting = (I40E_ITR_DYNAMIC |
 				       ITR_REG_TO_USEC(I40E_ITR_TX_DEF));
-	adapter->vsi.netdev = adapter->netdev;
-	adapter->vsi.qs_handle = adapter->vsi_res->qset_handle;
+	vsi->netdev = adapter->netdev;
+	vsi->qs_handle = adapter->vsi_res->qset_handle;
+	if (vfres->vf_offload_flags & I40E_VIRTCHNL_VF_OFFLOAD_RSS_PF) {
+		adapter->rss_key_size = vfres->rss_key_size;
+		adapter->rss_lut_size = vfres->rss_lut_size;
+	} else {
+		adapter->rss_key_size = I40EVF_HKEY_ARRAY_SIZE;
+		adapter->rss_lut_size = I40EVF_HLUT_ARRAY_SIZE;
+	}
+
 	return 0;
 }
 
@@ -2578,6 +2461,11 @@ static void i40evf_init_task(struct work_struct *work)
 	set_bit(__I40E_DOWN, &adapter->vsi.state);
 	i40evf_misc_irq_enable(adapter);
 
+	adapter->rss_key = kzalloc(adapter->rss_key_size, GFP_KERNEL);
+	adapter->rss_lut = kzalloc(adapter->rss_lut_size, GFP_KERNEL);
+	if (!adapter->rss_key || !adapter->rss_lut)
+		goto err_mem;
+
 	if (RSS_AQ(adapter)) {
 		adapter->aq_required |= I40EVF_FLAG_AQ_CONFIGURE_RSS;
 		mod_timer_pending(&adapter->watchdog_timer, jiffies + 1);
@@ -2588,7 +2476,8 @@ static void i40evf_init_task(struct work_struct *work)
 restart:
 	schedule_delayed_work(&adapter->init_task, msecs_to_jiffies(30));
 	return;
-
+err_mem:
+	i40evf_free_rss(adapter);
 err_register:
 	i40evf_free_misc_irq(adapter);
 err_sw_init:
@@ -2870,8 +2759,7 @@ static void i40evf_remove(struct pci_dev *pdev)
 
 	flush_scheduled_work();
 
-	/* Clear user configurations for RSS */
-	i40evf_clear_rss_config_user(&adapter->vsi);
+	i40evf_free_rss(adapter);
 
 	if (hw->aq.asq.count)
 		i40evf_shutdown_adminq(hw);
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index 488e738..e62c56b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -681,6 +681,115 @@ void i40evf_request_stats(struct i40evf_adapter *adapter)
 		/* if the request failed, don't lock out others */
 		adapter->current_op = I40E_VIRTCHNL_OP_UNKNOWN;
 }
+
+/**
+ * i40evf_get_hena
+ * @adapter: adapter structure
+ *
+ * Request hash enable capabilities from PF
+ **/
+void i40evf_get_hena(struct i40evf_adapter *adapter)
+{
+	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
+		/* bail because we already have a command pending */
+		dev_err(&adapter->pdev->dev, "Cannot get RSS hash capabilities, command %d pending\n",
+			adapter->current_op);
+		return;
+	}
+	adapter->current_op = I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS;
+	adapter->aq_required &= ~I40EVF_FLAG_AQ_GET_HENA;
+	i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS,
+			   NULL, 0);
+}
+
+/**
+ * i40evf_set_hena
+ * @adapter: adapter structure
+ *
+ * Request the PF to set our RSS hash capabilities
+ **/
+void i40evf_set_hena(struct i40evf_adapter *adapter)
+{
+	struct i40e_virtchnl_rss_hena vrh;
+
+	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
+		/* bail because we already have a command pending */
+		dev_err(&adapter->pdev->dev, "Cannot set RSS hash enable, command %d pending\n",
+			adapter->current_op);
+		return;
+	}
+	vrh.hena = adapter->hena;
+	adapter->current_op = I40E_VIRTCHNL_OP_SET_RSS_HENA;
+	adapter->aq_required &= ~I40EVF_FLAG_AQ_SET_HENA;
+	i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_SET_RSS_HENA,
+			   (u8 *)&vrh, sizeof(vrh));
+}
+
+/**
+ * i40evf_set_rss_key
+ * @adapter: adapter structure
+ *
+ * Request the PF to set our RSS hash key
+ **/
+void i40evf_set_rss_key(struct i40evf_adapter *adapter)
+{
+	struct i40e_virtchnl_rss_key *vrk;
+	int len;
+
+	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
+		/* bail because we already have a command pending */
+		dev_err(&adapter->pdev->dev, "Cannot set RSS key, command %d pending\n",
+			adapter->current_op);
+		return;
+	}
+	len = sizeof(struct i40e_virtchnl_rss_key) +
+	      (adapter->rss_key_size * sizeof(u8)) - 1;
+	vrk = kzalloc(len, GFP_KERNEL);
+	if (!vrk)
+		return;
+	vrk->vsi_id = adapter->vsi.id;
+	vrk->key_len = adapter->rss_key_size;
+	memcpy(vrk->key, adapter->rss_key, adapter->rss_key_size);
+
+	adapter->current_op = I40E_VIRTCHNL_OP_CONFIG_RSS_KEY;
+	adapter->aq_required &= ~I40EVF_FLAG_AQ_SET_RSS_KEY;
+	i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_CONFIG_RSS_KEY,
+			   (u8 *)vrk, len);
+	kfree(vrk);
+}
+
+/**
+ * i40evf_set_rss_lut
+ * @adapter: adapter structure
+ *
+ * Request the PF to set our RSS lookup table
+ **/
+void i40evf_set_rss_lut(struct i40evf_adapter *adapter)
+{
+	struct i40e_virtchnl_rss_lut *vrl;
+	int len;
+
+	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
+		/* bail because we already have a command pending */
+		dev_err(&adapter->pdev->dev, "Cannot set RSS LUT, command %d pending\n",
+			adapter->current_op);
+		return;
+	}
+	len = sizeof(struct i40e_virtchnl_rss_lut) +
+	      (adapter->rss_lut_size * sizeof(u8)) - 1;
+	vrl = kzalloc(len, GFP_KERNEL);
+	if (!vrl)
+		return;
+	vrl->vsi_id = adapter->vsi.id;
+	vrl->lut_entries = adapter->rss_lut_size;
+	memcpy(vrl->lut, adapter->rss_lut, adapter->rss_lut_size);
+	adapter->current_op = I40E_VIRTCHNL_OP_CONFIG_RSS_LUT;
+	adapter->aq_required &= ~I40EVF_FLAG_AQ_SET_RSS_LUT;
+	i40evf_send_pf_msg(adapter, I40E_VIRTCHNL_OP_CONFIG_RSS_LUT,
+			   (u8 *)vrl, len);
+	kfree(vrl);
+}
+
 /**
  * i40evf_request_reset
  * @adapter: adapter structure
@@ -820,6 +929,16 @@ void i40evf_virtchnl_completion(struct i40evf_adapter *adapter,
 		if (v_opcode != adapter->current_op)
 			return;
 		break;
+	case I40E_VIRTCHNL_OP_GET_RSS_HENA_CAPS: {
+		struct i40e_virtchnl_rss_hena *vrh =
+			(struct i40e_virtchnl_rss_hena *)msg;
+		if (msglen == sizeof(*vrh))
+			adapter->hena = vrh->hena;
+		else
+			dev_warn(&adapter->pdev->dev,
+				 "Invalid message %d from PF\n", v_opcode);
+		}
+		break;
 	default:
 		if (v_opcode != adapter->current_op)
 			dev_warn(&adapter->pdev->dev, "Expected response %d from PF, received %d\n",
-- 
2.5.5

[net-next 08/15] i40e: Allow user to change input set mask for flow director

[Jeff Kirsher]

From: Kiran Patil <kiran.patil@intel.com>

This patch implements feature, which allows user to change
input set mask for flow director using side-band channel.
This patch adds definition of FLOW_TYPE_MASK into the header file.
With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
src port, dst port) for flow type TCP4/UDP4.

Change-Id: I90052508f8c172c0da5a4b78b949704b4a59ea78
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h         |   6 ++
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 116 +++++++++++++++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_txrx.c    |   2 +-
 3 files changed, 123 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index e312adf..c2ac73d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -240,6 +240,12 @@ struct i40e_fdir_filter {
 	u32 fd_id;
 };
 
+#ifndef FLOW_TYPE_MASK
+#define FLOW_TYPE_MASK	0xFF
+#else
+#error FLOW_TYPE_MASK already defined elsewhere
+#endif
+
 #define I40E_ETH_P_LLDP			0x88cc
 
 #define I40E_DCB_PRIO_TYPE_STRICT	0
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 8e56c43..ada448b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2488,6 +2488,116 @@ static int i40e_del_fdir_entry(struct i40e_vsi *vsi,
 }
 
 /**
+ * i40e_handle_input_set - Detect and handle input set changes
+ * @vsi: pointer to the targeted VSI
+ * @fsp: pointer to RX flow classification spec
+ *
+ * Reads register, detect change in input set based on existing register
+ * value and what user has passed. Update input set mask register if needed.
+ **/
+static int i40e_handle_input_set(struct i40e_vsi *vsi,
+				 struct ethtool_rx_flow_spec *fsp)
+{
+	bool inset_mask_change = false;
+	struct i40e_pf *pf;
+	u64 val;
+	u8 idx;
+
+	if (unlikely(!vsi))
+		return -EINVAL;
+
+	pf = vsi->back;
+	switch (fsp->flow_type & FLOW_TYPE_MASK) {
+	case TCP_V4_FLOW:
+		idx = I40E_FILTER_PCTYPE_NONF_IPV4_TCP;
+		break;
+	case UDP_V4_FLOW:
+		idx = I40E_FILTER_PCTYPE_NONF_IPV4_UDP;
+		break;
+	default:
+		/* for all other flow types */
+		return 0;
+	}
+
+	val = ((u64)i40e_read_rx_ctl(&pf->hw,
+				     I40E_PRTQF_FD_INSET(idx, 1)) << 32) |
+	       (u64)i40e_read_rx_ctl(&pf->hw,
+				     I40E_PRTQF_FD_INSET(idx, 0));
+
+	/* Default input set (TCP/UDP/SCTP) contains following
+	 * fields: srcip + dest ip + src port + dest port
+	 * For SCTP, there is one extra field, "verification tag"
+	 */
+	if (val & I40E_L3_SRC_MASK) {
+		if (!fsp->h_u.tcp_ip4_spec.ip4src) {
+			val &= ~I40E_L3_SRC_MASK;
+			inset_mask_change = true;
+		}
+	} else {
+		if (fsp->h_u.tcp_ip4_spec.ip4src) {
+			val |= I40E_L3_SRC_MASK;
+			inset_mask_change = true;
+		}
+	}
+	if (val & I40E_L3_DST_MASK) {
+		if (!fsp->h_u.tcp_ip4_spec.ip4dst) {
+			val &= ~I40E_L3_DST_MASK;
+			inset_mask_change = true;
+		}
+	} else {
+		if (fsp->h_u.tcp_ip4_spec.ip4dst) {
+			val |= I40E_L3_DST_MASK;
+			inset_mask_change = true;
+		}
+	}
+	if (val & I40E_L4_SRC_MASK) {
+		if (!fsp->h_u.tcp_ip4_spec.psrc) {
+			val &= ~I40E_L4_SRC_MASK;
+			inset_mask_change = true;
+		}
+	} else {
+		if (fsp->h_u.tcp_ip4_spec.psrc) {
+			val |= I40E_L4_SRC_MASK;
+			inset_mask_change = true;
+		}
+	}
+	if (val & I40E_L4_DST_MASK) {
+		if (!fsp->h_u.tcp_ip4_spec.pdst) {
+			val &= ~I40E_L4_DST_MASK;
+			inset_mask_change = true;
+		}
+	} else {
+		if (fsp->h_u.tcp_ip4_spec.pdst) {
+			val |= I40E_L4_DST_MASK;
+			inset_mask_change = true;
+		}
+	}
+
+	if (inset_mask_change) {
+		if (pf->flags & I40E_FLAG_MFP_ENABLED) {
+			netif_err(pf, drv, vsi->netdev, "Change of input set is not supported when MFP mode is enabled\n");
+			return -EOPNOTSUPP;
+		}
+		if (pf->fdir_pf_active_filters) {
+			netif_err(pf, drv, vsi->netdev, "Change of input set is not supported when there are existing filters. Please delete them and re-try\n");
+			return -EOPNOTSUPP;
+		}
+
+		if (I40E_DEBUG_FD & pf->hw.debug_mask)
+			netif_info(pf, drv, vsi->netdev, "FD_INSET mask is changing to 0x%016llx\n",
+				   val);
+		/* Update input mask register since input set mask changed */
+		i40e_write_rx_ctl(&pf->hw, I40E_PRTQF_FD_INSET(idx, 1),
+				  (u32)(val >> 32));
+		i40e_write_rx_ctl(&pf->hw, I40E_PRTQF_FD_INSET(idx, 0),
+				  (u32)val);
+		netif_info(pf, drv, vsi->netdev, "Input set mask change has been successful. Please note that this and all other interfaces on (related and derived from) this part are affected as well.\n");
+	}
+
+	return 0;
+}
+
+/**
  * i40e_add_fdir_ethtool - Add/Remove Flow Director filters
  * @vsi: pointer to the targeted VSI
  * @cmd: command to get or set RX flow classification rules
@@ -2562,6 +2672,12 @@ static int i40e_add_fdir_ethtool(struct i40e_vsi *vsi,
 	input->dst_ip[0] = fsp->h_u.tcp_ip4_spec.ip4src;
 	input->src_ip[0] = fsp->h_u.tcp_ip4_spec.ip4dst;
 
+	ret = i40e_handle_input_set(vsi, fsp);
+	if (ret) {
+		netif_err(pf, drv, vsi->netdev, "Unable to handle change in input set mask\n");
+		goto free_input;
+	}
+
 	if (ntohl(fsp->m_ext.data[1])) {
 		vf_id = ntohl(fsp->h_ext.data[1]);
 		if (vf_id >= pf->num_alloc_vfs) {
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 6e44cf1..57af5f6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -412,7 +412,7 @@ int i40e_add_del_fdir(struct i40e_vsi *vsi,
 	struct i40e_pf *pf = vsi->back;
 	int ret;
 
-	switch (input->flow_type & ~FLOW_EXT) {
+	switch (input->flow_type & FLOW_TYPE_MASK) {
 	case TCP_V4_FLOW:
 		ret = i40e_add_del_fdir_tcpv4(vsi, input, add);
 		break;
-- 
2.5.5

[net-next 09/15] i40e: Add device capability which defines if update is available

[Jeff Kirsher]

From: Michal Kosiarz <michal.kosiarz@intel.com>

Add device capability which defines if update is available and security
check is needed during update process.

Change-ID: I380787c878275e1df18b39198df3ee3666342282
Signed-off-by: Michal Kosiarz <michal.kosiarz@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h   | 1 +
 drivers/net/ethernet/intel/i40e/i40e_common.c       | 6 ++++++
 drivers/net/ethernet/intel/i40e/i40e_type.h         | 5 +++++
 drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h | 1 +
 drivers/net/ethernet/intel/i40evf/i40e_type.h       | 5 +++++
 5 files changed, 18 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
index 8d5c65a..5179b3b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h
@@ -429,6 +429,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_SDP		0x0062
 #define I40E_AQ_CAP_ID_MDIO		0x0063
 #define I40E_AQ_CAP_ID_WSR_PROT		0x0064
+#define I40E_AQ_CAP_ID_NVM_MGMT		0x0080
 #define I40E_AQ_CAP_ID_FLEX10		0x00F1
 #define I40E_AQ_CAP_ID_CEM		0x00F2
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index f3c1d88..34e86f5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -3138,6 +3138,12 @@ static void i40e_parse_discover_capabilities(struct i40e_hw *hw, void *buff,
 			p->wr_csr_prot = (u64)number;
 			p->wr_csr_prot |= (u64)logical_id << 32;
 			break;
+		case I40E_AQ_CAP_ID_NVM_MGMT:
+			if (number & I40E_NVM_MGMT_SEC_REV_DISABLED)
+				p->sec_rev_disabled = true;
+			if (number & I40E_NVM_MGMT_UPDATE_DISABLED)
+				p->update_disabled = true;
+			break;
 		default:
 			break;
 		}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h
index bb57cd9..8aa14aa 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_type.h
@@ -275,6 +275,11 @@ struct i40e_hw_capabilities {
 #define I40E_FLEX10_STATUS_DCC_ERROR	0x1
 #define I40E_FLEX10_STATUS_VC_MODE	0x2
 
+	bool sec_rev_disabled;
+	bool update_disabled;
+#define I40E_NVM_MGMT_SEC_REV_DISABLED	0x1
+#define I40E_NVM_MGMT_UPDATE_DISABLED	0x2
+
 	bool mgmt_cem;
 	bool ieee_1588;
 	bool iwarp;
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
index aad8d62..1bcb8cf 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h
@@ -426,6 +426,7 @@ struct i40e_aqc_list_capabilities_element_resp {
 #define I40E_AQ_CAP_ID_SDP		0x0062
 #define I40E_AQ_CAP_ID_MDIO		0x0063
 #define I40E_AQ_CAP_ID_WSR_PROT		0x0064
+#define I40E_AQ_CAP_ID_NVM_MGMT		0x0080
 #define I40E_AQ_CAP_ID_FLEX10		0x00F1
 #define I40E_AQ_CAP_ID_CEM		0x00F2
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_type.h b/drivers/net/ethernet/intel/i40evf/i40e_type.h
index b720713..bfc97c2 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_type.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_type.h
@@ -258,6 +258,11 @@ struct i40e_hw_capabilities {
 #define I40E_FLEX10_STATUS_DCC_ERROR	0x1
 #define I40E_FLEX10_STATUS_VC_MODE	0x2
 
+	bool sec_rev_disabled;
+	bool update_disabled;
+#define I40E_NVM_MGMT_SEC_REV_DISABLED	0x1
+#define I40E_NVM_MGMT_UPDATE_DISABLED	0x2
+
 	bool mgmt_cem;
 	bool ieee_1588;
 	bool iwarp;
-- 
2.5.5

[net-next 10/15] i40e: Add DeviceID for X722 QSFP+

[Jeff Kirsher]

From: Kamil Krawczyk <kamil.krawczyk@intel.com>

Change-ID: I1370fbc7774e815ac1ad56561e97488e829592fc
Signed-off-by: Kamil Krawczyk <kamil.krawczyk@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_common.c   | 1 +
 drivers/net/ethernet/intel/i40e/i40e_devids.h   | 1 +
 drivers/net/ethernet/intel/i40evf/i40e_common.c | 1 +
 drivers/net/ethernet/intel/i40evf/i40e_devids.h | 1 +
 4 files changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 34e86f5..1db4790 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -61,6 +61,7 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_1G_BASE_T_X722:
 		case I40E_DEV_ID_10G_BASE_T_X722:
 		case I40E_DEV_ID_SFP_I_X722:
+		case I40E_DEV_ID_QSFP_I_X722:
 			hw->mac.type = I40E_MAC_X722;
 			break;
 		default:
diff --git a/drivers/net/ethernet/intel/i40e/i40e_devids.h b/drivers/net/ethernet/intel/i40e/i40e_devids.h
index dd4457d..d701861 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_devids.h
@@ -45,6 +45,7 @@
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
 #define I40E_DEV_ID_SFP_I_X722		0x37D3
+#define I40E_DEV_ID_QSFP_I_X722		0x37D4
 
 #define i40e_is_40G_device(d)		((d) == I40E_DEV_ID_QSFP_A  || \
 					 (d) == I40E_DEV_ID_QSFP_B  || \
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_common.c b/drivers/net/ethernet/intel/i40evf/i40e_common.c
index 4db0c03..8f64204 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_common.c
@@ -59,6 +59,7 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_1G_BASE_T_X722:
 		case I40E_DEV_ID_10G_BASE_T_X722:
 		case I40E_DEV_ID_SFP_I_X722:
+		case I40E_DEV_ID_QSFP_I_X722:
 			hw->mac.type = I40E_MAC_X722;
 			break;
 		case I40E_DEV_ID_X722_VF:
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_devids.h b/drivers/net/ethernet/intel/i40evf/i40e_devids.h
index 7023570..d34972b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_devids.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_devids.h
@@ -45,6 +45,7 @@
 #define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
 #define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
 #define I40E_DEV_ID_SFP_I_X722		0x37D3
+#define I40E_DEV_ID_QSFP_I_X722		0x37D4
 #define I40E_DEV_ID_X722_VF		0x37CD
 #define I40E_DEV_ID_X722_VF_HV		0x37D9
 
-- 
2.5.5

[net-next 11/15] i40e: Remove zero check

[Jeff Kirsher]

From: Greg Rose <gregory.v.rose@intel.com>

A mirror rule ID may be zero so do not return invalid parameter when the
user passes in a zero value for a rule ID.

Change-ID: I261b8c24725ce2c6ed32f859da81093dfcbe2970
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_common.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 1db4790..25872f2 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2668,10 +2668,7 @@ i40e_status i40e_aq_delete_mirrorrule(struct i40e_hw *hw, u16 sw_seid,
 			u16 *rules_used, u16 *rules_free)
 {
 	/* Rule ID has to be valid except rule_type: INGRESS VLAN mirroring */
-	if (rule_type != I40E_AQC_MIRROR_RULE_TYPE_VLAN) {
-		if (!rule_id)
-			return I40E_ERR_PARAM;
-	} else {
+	if (rule_type == I40E_AQC_MIRROR_RULE_TYPE_VLAN) {
 		/* count and mr_list shall be valid for rule_type INGRESS VLAN
 		 * mirroring. For other rule_type, count and rule_type should
 		 * not matter.
-- 
2.5.5

[net-next 12/15] i40e/i40evf: Only offload VLAN tag if enabled

[Jeff Kirsher]

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

The driver was offloading the VLAN tag into the skb
any time there was a VLAN tag and the hardware stripping was
enabled.  Just check to make sure it's enabled before put_tag.

Change-Id: Ife95290c06edd9a616393b38679923938b382241
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 3 ++-
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 57af5f6..be51a28 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1370,7 +1370,8 @@ static void i40e_receive_skb(struct i40e_ring *rx_ring,
 {
 	struct i40e_q_vector *q_vector = rx_ring->q_vector;
 
-	if (vlan_tag & VLAN_VID_MASK)
+	if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	    (vlan_tag & VLAN_VID_MASK))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
 
 	napi_gro_receive(&q_vector->napi, skb);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index f101895..4633235 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -842,7 +842,8 @@ static void i40e_receive_skb(struct i40e_ring *rx_ring,
 {
 	struct i40e_q_vector *q_vector = rx_ring->q_vector;
 
-	if (vlan_tag & VLAN_VID_MASK)
+	if ((rx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) &&
+	    (vlan_tag & VLAN_VID_MASK))
 		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tag);
 
 	napi_gro_receive(&q_vector->napi, skb);
-- 
2.5.5

[net-next 13/15] i40e: Add promiscuous on VLAN support

[Jeff Kirsher]

From: Greg Rose <gregory.v.rose@intel.com>

NFV use cases require the ability to steer packets to VSIs by VLAN tag
alone while being in promiscuous mode for multicast and unicast MAC
addresses.  These two new functions support that ability.

Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_common.c    | 70 ++++++++++++++++++++++++
 drivers/net/ethernet/intel/i40e/i40e_prototype.h |  8 +++
 2 files changed, 78 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_common.c b/drivers/net/ethernet/intel/i40e/i40e_common.c
index 25872f2..0e8552b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_common.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_common.c
@@ -2039,6 +2039,76 @@ i40e_status i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
 }
 
 /**
+ * i40e_aq_set_vsi_mc_promisc_on_vlan
+ * @hw: pointer to the hw struct
+ * @seid: vsi number
+ * @enable: set MAC L2 layer unicast promiscuous enable/disable for a given VLAN
+ * @vid: The VLAN tag filter - capture any multicast packet with this VLAN tag
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
+							 u16 seid, bool enable,
+							 u16 vid,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
+		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
+	enum i40e_status_code status;
+	u16 flags = 0;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_set_vsi_promiscuous_modes);
+
+	if (enable)
+		flags |= I40E_AQC_SET_VSI_PROMISC_MULTICAST;
+
+	cmd->promiscuous_flags = cpu_to_le16(flags);
+	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_MULTICAST);
+	cmd->seid = cpu_to_le16(seid);
+	cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID);
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
+ * i40e_aq_set_vsi_uc_promisc_on_vlan
+ * @hw: pointer to the hw struct
+ * @seid: vsi number
+ * @enable: set MAC L2 layer unicast promiscuous enable/disable for a given VLAN
+ * @vid: The VLAN tag filter - capture any unicast packet with this VLAN tag
+ * @cmd_details: pointer to command details structure or NULL
+ **/
+enum i40e_status_code i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw,
+							 u16 seid, bool enable,
+							 u16 vid,
+				struct i40e_asq_cmd_details *cmd_details)
+{
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_set_vsi_promiscuous_modes *cmd =
+		(struct i40e_aqc_set_vsi_promiscuous_modes *)&desc.params.raw;
+	enum i40e_status_code status;
+	u16 flags = 0;
+
+	i40e_fill_default_direct_cmd_desc(&desc,
+					  i40e_aqc_opc_set_vsi_promiscuous_modes);
+
+	if (enable)
+		flags |= I40E_AQC_SET_VSI_PROMISC_UNICAST;
+
+	cmd->promiscuous_flags = cpu_to_le16(flags);
+	cmd->valid_flags = cpu_to_le16(I40E_AQC_SET_VSI_PROMISC_UNICAST);
+	cmd->seid = cpu_to_le16(seid);
+	cmd->vlan_tag = cpu_to_le16(vid | I40E_AQC_SET_VSI_VLAN_VALID);
+
+	status = i40e_asq_send_command(hw, &desc, NULL, 0, cmd_details);
+
+	return status;
+}
+
+/**
  * i40e_aq_set_vsi_broadcast
  * @hw: pointer to the hw struct
  * @seid: vsi number
diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
index 134035f..8afb237 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h
@@ -133,6 +133,14 @@ i40e_status i40e_aq_set_vsi_unicast_promiscuous(struct i40e_hw *hw,
 		u16 vsi_id, bool set, struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_vsi_multicast_promiscuous(struct i40e_hw *hw,
 		u16 vsi_id, bool set, struct i40e_asq_cmd_details *cmd_details);
+enum i40e_status_code i40e_aq_set_vsi_mc_promisc_on_vlan(struct i40e_hw *hw,
+							 u16 seid, bool enable,
+							 u16 vid,
+				struct i40e_asq_cmd_details *cmd_details);
+enum i40e_status_code i40e_aq_set_vsi_uc_promisc_on_vlan(struct i40e_hw *hw,
+							 u16 seid, bool enable,
+							 u16 vid,
+				struct i40e_asq_cmd_details *cmd_details);
 i40e_status i40e_aq_set_vsi_vlan_promisc(struct i40e_hw *hw,
 				u16 seid, bool enable,
 				struct i40e_asq_cmd_details *cmd_details);
-- 
2.5.5

[net-next 14/15] i40e: Add VF promiscuous mode driver support

[Jeff Kirsher]

From: Anjali Singhai Jain <anjali.singhai@intel.com>

Add infrastructure for Network Function Virtualization VLAN tagged
packet steering feature.

Change-Id: I9b873d8fcc253858e6baba65ac68ec5b9363944e
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 153 ++++++++++++++++++++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h |   2 +
 2 files changed, 149 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index c364588..f47b0e8 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1427,6 +1427,25 @@ static void i40e_vc_reset_vf_msg(struct i40e_vf *vf)
 }
 
 /**
+ * i40e_getnum_vf_vsi_vlan_filters
+ * @vsi: pointer to the vsi
+ *
+ * called to get the number of VLANs offloaded on this VF
+ **/
+static inline int i40e_getnum_vf_vsi_vlan_filters(struct i40e_vsi *vsi)
+{
+	struct i40e_mac_filter *f;
+	int num_vlans = 0;
+
+	list_for_each_entry(f, &vsi->mac_filter_list, list) {
+		if (f->vlan >= 0 && f->vlan <= I40E_MAX_VLANID)
+			num_vlans++;
+	}
+
+	return num_vlans;
+}
+
+/**
  * i40e_vc_config_promiscuous_mode_msg
  * @vf: pointer to the VF info
  * @msg: pointer to the msg buffer
@@ -1442,22 +1461,122 @@ static int i40e_vc_config_promiscuous_mode_msg(struct i40e_vf *vf,
 	    (struct i40e_virtchnl_promisc_info *)msg;
 	struct i40e_pf *pf = vf->pf;
 	struct i40e_hw *hw = &pf->hw;
-	struct i40e_vsi *vsi;
+	struct i40e_mac_filter *f;
+	i40e_status aq_ret = 0;
 	bool allmulti = false;
-	i40e_status aq_ret;
+	struct i40e_vsi *vsi;
+	bool alluni = false;
+	int aq_err = 0;
 
 	vsi = i40e_find_vsi_from_id(pf, info->vsi_id);
 	if (!test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states) ||
 	    !test_bit(I40E_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps) ||
-	    !i40e_vc_isvalid_vsi_id(vf, info->vsi_id) ||
-	    (vsi->type != I40E_VSI_FCOE)) {
+	    !i40e_vc_isvalid_vsi_id(vf, info->vsi_id)) {
+		dev_err(&pf->pdev->dev,
+			"VF %d doesn't meet requirements to enter promiscuous mode\n",
+			vf->vf_id);
 		aq_ret = I40E_ERR_PARAM;
 		goto error_param;
 	}
+	/* Multicast promiscuous handling*/
 	if (info->flags & I40E_FLAG_VF_MULTICAST_PROMISC)
 		allmulti = true;
-	aq_ret = i40e_aq_set_vsi_multicast_promiscuous(hw, vsi->seid,
-						       allmulti, NULL);
+
+	if (vf->port_vlan_id) {
+		aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan(hw, vsi->seid,
+							    allmulti,
+							    vf->port_vlan_id,
+							    NULL);
+	} else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) {
+		list_for_each_entry(f, &vsi->mac_filter_list, list) {
+			if (f->vlan >= 0 && f->vlan <= I40E_MAX_VLANID)
+				aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan
+								   (hw,
+								   vsi->seid,
+								   allmulti,
+								   f->vlan,
+								   NULL);
+			aq_err = pf->hw.aq.asq_last_status;
+			if (aq_ret) {
+				dev_err(&pf->pdev->dev,
+					"Could not add VLAN %d to multicast promiscuous domain err %s aq_err %s\n",
+					f->vlan,
+					i40e_stat_str(&pf->hw, aq_ret),
+					i40e_aq_str(&pf->hw, aq_err));
+				break;
+			}
+		}
+	} else {
+		aq_ret = i40e_aq_set_vsi_multicast_promiscuous(hw, vsi->seid,
+							       allmulti, NULL);
+		aq_err = pf->hw.aq.asq_last_status;
+		if (aq_ret) {
+			dev_err(&pf->pdev->dev,
+				"VF %d failed to set multicast promiscuous mode err %s aq_err %s\n",
+				vf->vf_id,
+				i40e_stat_str(&pf->hw, aq_ret),
+				i40e_aq_str(&pf->hw, aq_err));
+			goto error_param_int;
+		}
+	}
+
+	if (!aq_ret) {
+		dev_info(&pf->pdev->dev,
+			 "VF %d successfully set multicast promiscuous mode\n",
+			 vf->vf_id);
+		if (allmulti)
+			set_bit(I40E_VF_STAT_MC_PROMISC, &vf->vf_states);
+		else
+			clear_bit(I40E_VF_STAT_MC_PROMISC, &vf->vf_states);
+	}
+
+	if (info->flags & I40E_FLAG_VF_UNICAST_PROMISC)
+		alluni = true;
+	if (vf->port_vlan_id) {
+		aq_ret = i40e_aq_set_vsi_uc_promisc_on_vlan(hw, vsi->seid,
+							    alluni,
+							    vf->port_vlan_id,
+							    NULL);
+	} else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) {
+		list_for_each_entry(f, &vsi->mac_filter_list, list) {
+			aq_ret = 0;
+			if (f->vlan >= 0 && f->vlan <= I40E_MAX_VLANID)
+				aq_ret =
+				i40e_aq_set_vsi_uc_promisc_on_vlan(hw,
+								   vsi->seid,
+								   alluni,
+								   f->vlan,
+								   NULL);
+				aq_err = pf->hw.aq.asq_last_status;
+			if (aq_ret)
+				dev_err(&pf->pdev->dev,
+					"Could not add VLAN %d to Unicast promiscuous domain err %s aq_err %s\n",
+					f->vlan,
+					i40e_stat_str(&pf->hw, aq_ret),
+					i40e_aq_str(&pf->hw, aq_err));
+		}
+	} else {
+		aq_ret = i40e_aq_set_vsi_unicast_promiscuous(hw, vsi->seid,
+							     allmulti, NULL);
+		aq_err = pf->hw.aq.asq_last_status;
+		if (aq_ret)
+			dev_err(&pf->pdev->dev,
+				"VF %d failed to set unicast promiscuous mode %8.8x err %s aq_err %s\n",
+				vf->vf_id, info->flags,
+				i40e_stat_str(&pf->hw, aq_ret),
+				i40e_aq_str(&pf->hw, aq_err));
+	}
+
+error_param_int:
+	if (!aq_ret) {
+		dev_info(&pf->pdev->dev,
+			 "VF %d successfully set unicast promiscuous mode\n",
+			 vf->vf_id);
+		if (alluni)
+			set_bit(I40E_VF_STAT_UC_PROMISC, &vf->vf_states);
+		else
+			clear_bit(I40E_VF_STAT_UC_PROMISC, &vf->vf_states);
+	}
 
 error_param:
 	/* send the response to the VF */
@@ -1919,6 +2038,17 @@ static int i40e_vc_add_vlan_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 		/* add new VLAN filter */
 		int ret = i40e_vsi_add_vlan(vsi, vfl->vlan_id[i]);
 
+		if (test_bit(I40E_VF_STAT_UC_PROMISC, &vf->vf_states))
+			i40e_aq_set_vsi_uc_promisc_on_vlan(&pf->hw, vsi->seid,
+							   true,
+							   vfl->vlan_id[i],
+							   NULL);
+		if (test_bit(I40E_VF_STAT_MC_PROMISC, &vf->vf_states))
+			i40e_aq_set_vsi_mc_promisc_on_vlan(&pf->hw, vsi->seid,
+							   true,
+							   vfl->vlan_id[i],
+							   NULL);
+
 		if (ret)
 			dev_err(&pf->pdev->dev,
 				"Unable to add VLAN filter %d for VF %d, error %d\n",
@@ -1971,6 +2101,17 @@ static int i40e_vc_remove_vlan_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	for (i = 0; i < vfl->num_elements; i++) {
 		int ret = i40e_vsi_kill_vlan(vsi, vfl->vlan_id[i]);
 
+		if (test_bit(I40E_VF_STAT_UC_PROMISC, &vf->vf_states))
+			i40e_aq_set_vsi_uc_promisc_on_vlan(&pf->hw, vsi->seid,
+							   false,
+							   vfl->vlan_id[i],
+							   NULL);
+		if (test_bit(I40E_VF_STAT_MC_PROMISC, &vf->vf_states))
+			i40e_aq_set_vsi_mc_promisc_on_vlan(&pf->hw, vsi->seid,
+							   false,
+							   vfl->vlan_id[i],
+							   NULL);
+
 		if (ret)
 			dev_err(&pf->pdev->dev,
 				"Unable to delete VLAN filter %d for VF %d, error %d\n",
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
index 838cbd2..8cbf579 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.h
@@ -61,6 +61,8 @@ enum i40e_vf_states {
 	I40E_VF_STAT_IWARPENA,
 	I40E_VF_STAT_FCOEENA,
 	I40E_VF_STAT_DISABLED,
+	I40E_VF_STAT_MC_PROMISC,
+	I40E_VF_STAT_UC_PROMISC,
 };
 
 /* VF capabilities */
-- 
2.5.5

[net-next 15/15] i40evf: Add driver support for promiscuous mode

[Jeff Kirsher]

From: Anjali Singhai Jain <anjali.singhai@intel.com>

Add necessary Linux Ethernet driver support for promiscuous mode
operation. Add a flag so the VF knows it is in promiscuous mode
and two state flags to discreetly track multicast and unicast
promiscuous states.

Change-Id: Ib2f2dc7a7582304fec90fc917ebb7ded21ba1de4
Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Greg Rose <gregory.v.rose@intel.com>
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c  | 14 +++++++-------
 drivers/net/ethernet/intel/i40evf/i40evf.h          |  3 +++
 drivers/net/ethernet/intel/i40evf/i40evf_main.c     | 19 +++++++++++++++++++
 drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c | 11 +++++++++++
 4 files changed, 40 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index f47b0e8..c226c2d 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1489,13 +1489,13 @@ static int i40e_vc_config_promiscuous_mode_msg(struct i40e_vf *vf,
 							    NULL);
 	} else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) {
 		list_for_each_entry(f, &vsi->mac_filter_list, list) {
-			if (f->vlan >= 0 && f->vlan <= I40E_MAX_VLANID)
-				aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan
-								   (hw,
-								   vsi->seid,
-								   allmulti,
-								   f->vlan,
-								   NULL);
+			if (f->vlan < 0 || f->vlan > I40E_MAX_VLANID)
+				continue;
+			aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan(hw,
+								    vsi->seid,
+								    allmulti,
+								    f->vlan,
+								    NULL);
 			aq_err = pf->hw.aq.asq_last_status;
 			if (aq_ret) {
 				dev_err(&pf->pdev->dev,
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index 63f7aae..25afabf 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -220,6 +220,7 @@ struct i40evf_adapter {
 #define I40EVF_FLAG_WB_ON_ITR_CAPABLE		BIT(11)
 #define I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE	BIT(12)
 #define I40EVF_FLAG_ADDR_SET_BY_PF		BIT(13)
+#define I40EVF_FLAG_PROMISC_ON			BIT(15)
 /* duplicates for common code */
 #define I40E_FLAG_FDIR_ATR_ENABLED		 0
 #define I40E_FLAG_DCB_ENABLED			 0
@@ -244,6 +245,8 @@ struct i40evf_adapter {
 #define I40EVF_FLAG_AQ_SET_HENA			BIT(12)
 #define I40EVF_FLAG_AQ_SET_RSS_KEY		BIT(13)
 #define I40EVF_FLAG_AQ_SET_RSS_LUT		BIT(14)
+#define I40EVF_FLAG_AQ_REQUEST_PROMISC		BIT(15)
+#define I40EVF_FLAG_AQ_RELEASE_PROMISC		BIT(16)
 
 	/* OS defined structs */
 	struct net_device *netdev;
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index af53159..d1c4afd 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -943,6 +943,14 @@ static void i40evf_set_rx_mode(struct net_device *netdev)
 bottom_of_search_loop:
 		continue;
 	}
+
+	if (netdev->flags & IFF_PROMISC &&
+	    !(adapter->flags & I40EVF_FLAG_PROMISC_ON))
+		adapter->aq_required |= I40EVF_FLAG_AQ_REQUEST_PROMISC;
+	else if (!(netdev->flags & IFF_PROMISC) &&
+		 adapter->flags & I40EVF_FLAG_PROMISC_ON)
+		adapter->aq_required |= I40EVF_FLAG_AQ_RELEASE_PROMISC;
+
 	clear_bit(__I40EVF_IN_CRITICAL_TASK, &adapter->crit_section);
 }
 
@@ -1622,6 +1630,17 @@ static void i40evf_watchdog_task(struct work_struct *work)
 		goto watchdog_done;
 	}
 
+	if (adapter->aq_required & I40EVF_FLAG_AQ_REQUEST_PROMISC) {
+		i40evf_set_promiscuous(adapter, I40E_FLAG_VF_UNICAST_PROMISC |
+				       I40E_FLAG_VF_MULTICAST_PROMISC);
+		goto watchdog_done;
+	}
+
+	if (adapter->aq_required & I40EVF_FLAG_AQ_RELEASE_PROMISC) {
+		i40evf_set_promiscuous(adapter, 0);
+		goto watchdog_done;
+	}
+
 	if (adapter->state == __I40EVF_RUNNING)
 		i40evf_request_stats(adapter);
 watchdog_done:
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index e62c56b..ba7fbc0 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -652,6 +652,17 @@ void i40evf_set_promiscuous(struct i40evf_adapter *adapter, int flags)
 			adapter->current_op);
 		return;
 	}
+
+	if (flags) {
+		adapter->flags |= I40EVF_FLAG_PROMISC_ON;
+		adapter->aq_required &= ~I40EVF_FLAG_AQ_REQUEST_PROMISC;
+		dev_info(&adapter->pdev->dev, "Entering promiscuous mode\n");
+	} else {
+		adapter->flags &= ~I40EVF_FLAG_PROMISC_ON;
+		adapter->aq_required &= ~I40EVF_FLAG_AQ_RELEASE_PROMISC;
+		dev_info(&adapter->pdev->dev, "Leaving promiscuous mode\n");
+	}
+
 	adapter->current_op = I40E_VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE;
 	vpi.vsi_id = adapter->vsi_res->vsi_id;
 	vpi.flags = flags;
-- 
2.5.5

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Tue, 26 Apr 2016 13:55:41 -0700

> From: Kiran Patil <kiran.patil@intel.com>
> 
> This patch implements feature, which allows user to change
> input set mask for flow director using side-band channel.
> This patch adds definition of FLOW_TYPE_MASK into the header file.
> With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
> src port, dst port) for flow type TCP4/UDP4.
> 
> Change-Id: I90052508f8c172c0da5a4b78b949704b4a59ea78
> Signed-off-by: Kiran Patil <kiran.patil@intel.com>
> Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

If you want to do this, you have to define the semantics generically
in include/uapi/linux/ethtool.h so that other drivers can implement
this too.

Please don't ever implement private, driver specific, interpretations
of the generic ethtool facilitites.

The semantics and interpretations of the values must absolutely be
consistent across all drivers in the tree.

Otherwise the user experience is terrible.

Thanks.

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[Patil, Kiran]

On 4/26/2016 8:48 PM, David Miller wrote:
> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> Date: Tue, 26 Apr 2016 13:55:41 -0700
>
>> From: Kiran Patil <kiran.patil@intel.com>
>>
>> This patch implements feature, which allows user to change
>> input set mask for flow director using side-band channel.
>> This patch adds definition of FLOW_TYPE_MASK into the header file.
>> With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
>> src port, dst port) for flow type TCP4/UDP4.
>>
>> Change-Id: I90052508f8c172c0da5a4b78b949704b4a59ea78
>> Signed-off-by: Kiran Patil <kiran.patil@intel.com>
>> Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
>> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>
> If you want to do this, you have to define the semantics generically
> in include/uapi/linux/ethtool.h so that other drivers can implement
> this too.
>
> Please don't ever implement private, driver specific, interpretations
> of the generic ethtool facilitites.
>
> The semantics and interpretations of the values must absolutely be
> consistent across all drivers in the tree.
>
> Otherwise the user experience is terrible.
>
> Thanks.
>
This is not new feature implemented in i40e driver. This is the original 
feature of ethtool which allows user to specify subset of tuple for 
setting up flow director.

i40e driver using it same way as ixgbe.

Please let us know if I misinterpreted your response.

Would you recommend that we re-submit patch with better patch 
description (indicating that it is not new feature but just enabling 
feature).

Thanks,
-- Kiran P.

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[John Fastabend]

On 16-05-04 07:47 PM, Patil, Kiran wrote:
> On 4/26/2016 8:48 PM, David Miller wrote:
>> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>> Date: Tue, 26 Apr 2016 13:55:41 -0700
>>
>>> From: Kiran Patil <kiran.patil@intel.com>
>>>
>>> This patch implements feature, which allows user to change
>>> input set mask for flow director using side-band channel.
>>> This patch adds definition of FLOW_TYPE_MASK into the header file.
>>> With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
>>> src port, dst port) for flow type TCP4/UDP4.
>>>
>>> Change-Id: I90052508f8c172c0da5a4b78b949704b4a59ea78
>>> Signed-off-by: Kiran Patil <kiran.patil@intel.com>
>>> Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
>>> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>>
>> If you want to do this, you have to define the semantics generically
>> in include/uapi/linux/ethtool.h so that other drivers can implement
>> this too.
>>
>> Please don't ever implement private, driver specific, interpretations
>> of the generic ethtool facilitites.
>>
>> The semantics and interpretations of the values must absolutely be
>> consistent across all drivers in the tree.
>>
>> Otherwise the user experience is terrible.
>>
>> Thanks.
>>
> This is not new feature implemented in i40e driver. This is the original
> feature of ethtool which allows user to specify subset of tuple for
> setting up flow director.
> 
> i40e driver using it same way as ixgbe.
> 
> Please let us know if I misinterpreted your response.
> 
> Would you recommend that we re-submit patch with better patch
> description (indicating that it is not new feature but just enabling
> feature).
> 
> Thanks,
> -- Kiran P.

At least define FLOW_TYPE_MASK in ethtool.h then its not
sort of hobbled together in the driver and others can use it
instead of the normal ~FLOW_EXT which I see other drivers used.
Another benefit if its near the definition of the flow types
you have a chance of someone seeing it if they add a flag
past 0xff.

And maybe do it as a separate patch. So you aren't adding normal
driver ethtool implementation and a new #define for all drivers
in the same patch.

.John

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[Alexander Duyck]

On Tue, Apr 26, 2016 at 1:55 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> From: Kiran Patil <kiran.patil@intel.com>
>
> This patch implements feature, which allows user to change
> input set mask for flow director using side-band channel.
> This patch adds definition of FLOW_TYPE_MASK into the header file.
> With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
> src port, dst port) for flow type TCP4/UDP4.

Isn't that what the mask field in the flow specifier is for?  It seems
like you guys are trying to come up with your own alternative to how
this has been done in the past.  It looks like this driver is badly
broken and it isn't implementing flow director.  It is implementing a
bad hack that is trying to pass itself off as flow director.

I don't know who thought it was okay to come up with your own
interpretation on how the fields are supposed to be used but whoever
did it made a real mess of things.  I am pretty sure the ixgbe driver
and the i40e driver are playing by two completely different rule
books.  If you printed an i40e rule and tried to apply it to ixgbe it
just wouldn't work because you guys didn't do basic things like
actually use the mask fields, in the other direction you guys totally
ignored the mask fields which really isn't okay because you should
throw an error if someone tries to get the driver to do something it
cannot support.

I'd say you are better off scrapping this whole patch and fixing flow
director so that you actually use the mask fields correctly because
the current implementation is badly broken and breaking it further
isn't going to help.  You need to add support for validating the mask
fields to the driver and that will solve the original problem that
this patch is attempting to work around.

- Alex

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[David Miller]

From: "Patil, Kiran" <kiran.patil@intel.com>
Date: Wed, 4 May 2016 19:47:26 -0700

> This is not new feature implemented in i40e driver. This is the
> original feature of ethtool which allows user to specify subset of
> tuple for setting up flow director.

Where is the interpretation of this special value defined in the
ethtool headers?

If it doesn't even have a macro for that magic constant value, it's
not part of the interface at all.

If the only other example of doing this trick is another Intel driver,
that lends even less credence to your arguments that this is a defined
interface.  Rather, it makes it look even more like an Intel specific
behavior.  Vendor specific behaviors of interfaces is exactly what we
are trying to avoid.

Re: [net-next 08/15] i40e: Allow user to change input set mask for flow director

[Patil, Kiran]

On 5/4/2016 8:22 PM, John Fastabend wrote:
> On 16-05-04 07:47 PM, Patil, Kiran wrote:
>> On 4/26/2016 8:48 PM, David Miller wrote:
>>> From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>>> Date: Tue, 26 Apr 2016 13:55:41 -0700
>>>
>>>> From: Kiran Patil <kiran.patil@intel.com>
>>>>
>>>> This patch implements feature, which allows user to change
>>>> input set mask for flow director using side-band channel.
>>>> This patch adds definition of FLOW_TYPE_MASK into the header file.
>>>> With this patch, user can now specify less than 4 tuple(src ip, dsp ip,
>>>> src port, dst port) for flow type TCP4/UDP4.
>>>>
>>>> Change-Id: I90052508f8c172c0da5a4b78b949704b4a59ea78
>>>> Signed-off-by: Kiran Patil <kiran.patil@intel.com>
>>>> Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
>>>> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
>>>
>>> If you want to do this, you have to define the semantics generically
>>> in include/uapi/linux/ethtool.h so that other drivers can implement
>>> this too.
>>>
>>> Please don't ever implement private, driver specific, interpretations
>>> of the generic ethtool facilitites.
>>>
>>> The semantics and interpretations of the values must absolutely be
>>> consistent across all drivers in the tree.
>>>
>>> Otherwise the user experience is terrible.
>>>
>>> Thanks.
>>>
>> This is not new feature implemented in i40e driver. This is the original
>> feature of ethtool which allows user to specify subset of tuple for
>> setting up flow director.
>>
>> i40e driver using it same way as ixgbe.
>>
>> Please let us know if I misinterpreted your response.
>>
>> Would you recommend that we re-submit patch with better patch
>> description (indicating that it is not new feature but just enabling
>> feature).
>>
>> Thanks,
>> -- Kiran P.
>
> At least define FLOW_TYPE_MASK in ethtool.h then its not
> sort of hobbled together in the driver and others can use it
> instead of the normal ~FLOW_EXT which I see other drivers used.
> Another benefit if its near the definition of the flow types
> you have a chance of someone seeing it if they add a flag
> past 0xff.
>
> And maybe do it as a separate patch. So you aren't adding normal
> driver ethtool implementation and a new #define for all drivers
> in the same patch.
>
> .John
>
Now I see what the problem is (FLOW_TYPE_MASK). Unfortunately it was 
never meant to be part of this patch since it was introduced by other 
patch (Cloud filter support) and was suggested to move into this header 
file. As suggested, we can create another patch just for this 
FLOW_TYPE_MASK in ethtool.h and re-work original patch to not have 
define for FLOW_TYPE_MASK.

May be we can go back and think, it this define absolutely needed or not.

Thanks,
-- Kiran P.