i40e/i40evf: Add TX/RX outer UDP checksum support for X722

X722 supports offloading of outer UDP TX and RX checksum for tunneled
packets. This patch exposes the support and leaves it enabled by
default.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Jim Young <james.m.young@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

drivers/net/ethernet/intel/i40e/i40e_main.c

@@ -7073,6 +7073,8 @@ static int i40e_alloc_rings(struct i40e_vsi *vsi)
 		tx_ring->dcb_tc = 0;
 		if (vsi->back->flags & I40E_FLAG_WB_ON_ITR_CAPABLE)
 			tx_ring->flags = I40E_TXR_FLAGS_WB_ON_ITR;
+		if (vsi->back->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE)
+			tx_ring->flags |= I40E_TXR_FLAGS_OUTER_UDP_CSUM;
 		vsi->tx_rings[i] = tx_ring;
 
 		rx_ring = &tx_ring[1];

drivers/net/ethernet/intel/i40e/i40e_txrx.c

@@ -1429,7 +1429,8 @@ static inline void i40e_rx_checksum(struct i40e_vsi *vsi,
 	 * so the total length of IPv4 header is IHL*4 bytes
 	 * The UDP_0 bit *may* bet set if the *inner* header is UDP
 	 */
-	if (ipv4_tunnel) {
+	if (!(vsi->back->flags & I40E_FLAG_OUTER_UDP_CSUM_CAPABLE) &&
+	    (ipv4_tunnel)) {
 		skb->transport_header = skb->mac_header +
 					sizeof(struct ethhdr) +
 					(ip_hdr(skb)->ihl * 4);
@@ -2301,11 +2302,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	struct iphdr *this_ip_hdr;
 	u32 network_hdr_len;
 	u8 l4_hdr = 0;
+	struct udphdr *oudph;
+	struct iphdr *oiph;
 	u32 l4_tunnel = 0;
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
 		case IPPROTO_UDP:
+			oudph = udp_hdr(skb);
+			oiph = ip_hdr(skb);
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
@@ -2342,6 +2347,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags &= ~I40E_TX_FLAGS_IPV4;
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
+		if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) &&
+		    (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING)        &&
+		    (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) {
+			oudph->check = ~csum_tcpudp_magic(oiph->saddr,
+					oiph->daddr,
+					(skb->len - skb_transport_offset(skb)),
+					IPPROTO_UDP, 0);
+			*cd_tunneling |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
+		}
 	} else {
 		network_hdr_len = skb_network_header_len(skb);
 		this_ip_hdr = ip_hdr(skb);

drivers/net/ethernet/intel/i40e/i40e_txrx.h

@@ -267,6 +267,8 @@ struct i40e_ring {
 
 	u16 flags;
 #define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
+#define I40E_TXR_FLAGS_OUTER_UDP_CSUM	BIT(1)
+
 	/* stats structs */
 	struct i40e_queue_stats	stats;
 	struct u64_stats_sync syncp;

drivers/net/ethernet/intel/i40e/i40e_type.h

@@ -607,14 +607,18 @@ enum i40e_rx_desc_status_bits {
 	I40E_RX_DESC_STATUS_CRCP_SHIFT		= 4,
 	I40E_RX_DESC_STATUS_TSYNINDX_SHIFT	= 5, /* 2 BITS */
 	I40E_RX_DESC_STATUS_TSYNVALID_SHIFT	= 7,
-	I40E_RX_DESC_STATUS_PIF_SHIFT		= 8,
+	/* Note: Bit 8 is reserved in X710 and XL710 */
+	I40E_RX_DESC_STATUS_EXT_UDP_0_SHIFT	= 8,
 	I40E_RX_DESC_STATUS_UMBCAST_SHIFT	= 9, /* 2 BITS */
 	I40E_RX_DESC_STATUS_FLM_SHIFT		= 11,
 	I40E_RX_DESC_STATUS_FLTSTAT_SHIFT	= 12, /* 2 BITS */
 	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14,
 	I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT	= 15,
 	I40E_RX_DESC_STATUS_RESERVED_SHIFT	= 16, /* 2 BITS */
-	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18,
+	/* Note: For non-tunnel packets INT_UDP_0 is the right status for
+	 * UDP header
+	 */
+	I40E_RX_DESC_STATUS_INT_UDP_0_SHIFT	= 18,
 	I40E_RX_DESC_STATUS_LAST /* this entry must be last!!! */
 };
 
@@ -955,6 +959,8 @@ enum i40e_tx_ctx_desc_eipt_offload {
 #define I40E_TXD_CTX_QW0_DECTTL_MASK	(0xFULL << \
 					 I40E_TXD_CTX_QW0_DECTTL_SHIFT)
 
+#define I40E_TXD_CTX_QW0_L4T_CS_SHIFT	23
+#define I40E_TXD_CTX_QW0_L4T_CS_MASK	BIT_ULL(I40E_TXD_CTX_QW0_L4T_CS_SHIFT)
 struct i40e_filter_program_desc {
 	__le32 qindex_flex_ptype_vsi;
 	__le32 rsvd;

drivers/net/ethernet/intel/i40evf/i40e_txrx.c

@@ -1528,11 +1528,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 	struct iphdr *this_ip_hdr;
 	u32 network_hdr_len;
 	u8 l4_hdr = 0;
+	struct udphdr *oudph;
+	struct iphdr *oiph;
 	u32 l4_tunnel = 0;
 
 	if (skb->encapsulation) {
 		switch (ip_hdr(skb)->protocol) {
 		case IPPROTO_UDP:
+			oudph = udp_hdr(skb);
+			oiph = ip_hdr(skb);
 			l4_tunnel = I40E_TXD_CTX_UDP_TUNNELING;
 			*tx_flags |= I40E_TX_FLAGS_VXLAN_TUNNEL;
 			break;
@@ -1571,6 +1575,15 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 		}
 
 
+		if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) &&
+		    (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING)        &&
+		    (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) {
+			oudph->check = ~csum_tcpudp_magic(oiph->saddr,
+					oiph->daddr,
+					(skb->len - skb_transport_offset(skb)),
+					IPPROTO_UDP, 0);
+			*cd_tunneling |= I40E_TXD_CTX_QW0_L4T_CS_MASK;
+		}
 	} else {
 		network_hdr_len = skb_network_header_len(skb);
 		this_ip_hdr = ip_hdr(skb);

drivers/net/ethernet/intel/i40evf/i40e_txrx.h

@@ -264,6 +264,8 @@ struct i40e_ring {
 
 	u16 flags;
 #define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
+#define I40E_TXR_FLAGS_OUTER_UDP_CSUM	BIT(1)
+
 	/* stats structs */
 	struct i40e_queue_stats	stats;
 	struct u64_stats_sync syncp;

drivers/net/ethernet/intel/i40evf/i40e_type.h

@@ -601,14 +601,18 @@ enum i40e_rx_desc_status_bits {
 	I40E_RX_DESC_STATUS_CRCP_SHIFT		= 4,
 	I40E_RX_DESC_STATUS_TSYNINDX_SHIFT	= 5, /* 2 BITS */
 	I40E_RX_DESC_STATUS_TSYNVALID_SHIFT	= 7,
-	I40E_RX_DESC_STATUS_PIF_SHIFT		= 8,
+	/* Note: Bit 8 is reserved in X710 and XL710 */
+	I40E_RX_DESC_STATUS_EXT_UDP_0_SHIFT	= 8,
 	I40E_RX_DESC_STATUS_UMBCAST_SHIFT	= 9, /* 2 BITS */
 	I40E_RX_DESC_STATUS_FLM_SHIFT		= 11,
 	I40E_RX_DESC_STATUS_FLTSTAT_SHIFT	= 12, /* 2 BITS */
 	I40E_RX_DESC_STATUS_LPBK_SHIFT		= 14,
 	I40E_RX_DESC_STATUS_IPV6EXADD_SHIFT	= 15,
 	I40E_RX_DESC_STATUS_RESERVED_SHIFT	= 16, /* 2 BITS */
-	I40E_RX_DESC_STATUS_UDP_0_SHIFT		= 18,
+	/* Note: For non-tunnel packets INT_UDP_0 is the right status for
+	 * UDP header
+	 */
+	I40E_RX_DESC_STATUS_INT_UDP_0_SHIFT	= 18,
 	I40E_RX_DESC_STATUS_LAST /* this entry must be last!!! */
 };
 
@@ -949,6 +953,8 @@ enum i40e_tx_ctx_desc_eipt_offload {
 #define I40E_TXD_CTX_QW0_DECTTL_MASK	(0xFULL << \
 					 I40E_TXD_CTX_QW0_DECTTL_SHIFT)
 
+#define I40E_TXD_CTX_QW0_L4T_CS_SHIFT	23
+#define I40E_TXD_CTX_QW0_L4T_CS_MASK	BIT_ULL(I40E_TXD_CTX_QW0_L4T_CS_SHIFT)
 struct i40e_filter_program_desc {
 	__le32 qindex_flex_ptype_vsi;
 	__le32 rsvd;