Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue

Jeff Kirsher says:

====================
Intel Wired LAN Driver Updates 2015-08-05

This series contains updates to i40e, i40evf and e1000e.

Anjali adds support for x772 devices to i40e and i40evf.  With the added
support, x772 supports offloading of the outer UDP transmit and receive
checksum for tunneled packets.  Also supports evicting ATR filters in the
hardware, so update the driver with this new feature set.

Raanan provides several fixes for e1000e, first rectifies the Energy
Efficient Ethernet in Sx code so that it only applies to parts that
actually support EEE in Sx.  Fix whitespace and moved ICH8 related define
to the proper context.  Fixed the ASPM locking which was reported by
Bjorn Helgaas.  Fix a workaround implementation for systime which could
experience a large non-linear increment of the systime value when
checking for overflow.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/intel/e1000e/ich8lan.h

@@ -113,7 +113,7 @@
 #define NVM_SIZE_MULTIPLIER 4096	/*multiplier for NVMS field */
 #define E1000_FLASH_BASE_ADDR 0xE000	/*offset of NVM access regs */
 #define E1000_CTRL_EXT_NVMVS 0x3	/*NVM valid sector */
-
+#define E1000_TARC0_CB_MULTIQ_3_REQ	(1 << 28 | 1 << 29)
 #define PCIE_ICH8_SNOOP_ALL	PCIE_NO_SNOOP_ALL
 
 #define E1000_ICH_RAR_ENTRIES	7

drivers/net/ethernet/intel/e1000e/netdev.c

@@ -48,7 +48,7 @@
 
 #define DRV_EXTRAVERSION "-k"
 
-#define DRV_VERSION "3.2.5" DRV_EXTRAVERSION
+#define DRV_VERSION "3.2.6" DRV_EXTRAVERSION
 char e1000e_driver_name[] = "e1000e";
 const char e1000e_driver_version[] = DRV_VERSION;
 
@@ -4280,18 +4280,29 @@ static cycle_t e1000e_cyclecounter_read(const struct cyclecounter *cc)
 	struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
 						     cc);
 	struct e1000_hw *hw = &adapter->hw;
+	u32 systimel_1, systimel_2, systimeh;
 	cycle_t systim, systim_next;
-	/* SYSTIMH latching upon SYSTIML read does not work well. To fix that
-	 * we don't want to allow overflow of SYSTIML and a change to SYSTIMH
-	 * to occur between reads, so if we read a vale close to overflow, we
-	 * wait for overflow to occur and read both registers when its safe.
+	/* SYSTIMH latching upon SYSTIML read does not work well.
+	 * This means that if SYSTIML overflows after we read it but before
+	 * we read SYSTIMH, the value of SYSTIMH has been incremented and we
+	 * will experience a huge non linear increment in the systime value
+	 * to fix that we test for overflow and if true, we re-read systime.
+	 */
+	systimel_1 = er32(SYSTIML);
+	systimeh = er32(SYSTIMH);
+	systimel_2 = er32(SYSTIML);
+	/* Check for overflow. If there was no overflow, use the values */
+	if (systimel_1 < systimel_2) {
+		systim = (cycle_t)systimel_1;
+		systim |= (cycle_t)systimeh << 32;
+	} else {
+		/* There was an overflow, read again SYSTIMH, and use
+		 * systimel_2
 		 */
-	u32 systim_overflow_latch_fix = 0x3FFFFFFF;
-
-	do {
-		systim = (cycle_t)er32(SYSTIML);
-	} while (systim > systim_overflow_latch_fix);
-	systim |= (cycle_t)er32(SYSTIMH) << 32;
+		systimeh = er32(SYSTIMH);
+		systim = (cycle_t)systimel_2;
+		systim |= (cycle_t)systimeh << 32;
+	}
 
 	if ((hw->mac.type == e1000_82574) || (hw->mac.type == e1000_82583)) {
 		u64 incvalue, time_delta, rem, temp;
@@ -6317,6 +6328,33 @@ static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
 			return retval;
 	}
 
+	/* Ensure that the appropriate bits are set in LPI_CTRL
+	 * for EEE in Sx
+	 */
+	if ((hw->phy.type >= e1000_phy_i217) &&
+	    adapter->eee_advert && hw->dev_spec.ich8lan.eee_lp_ability) {
+		u16 lpi_ctrl = 0;
+
+		retval = hw->phy.ops.acquire(hw);
+		if (!retval) {
+			retval = e1e_rphy_locked(hw, I82579_LPI_CTRL,
+						 &lpi_ctrl);
+			if (!retval) {
+				if (adapter->eee_advert &
+				    hw->dev_spec.ich8lan.eee_lp_ability &
+				    I82579_EEE_100_SUPPORTED)
+					lpi_ctrl |= I82579_LPI_CTRL_100_ENABLE;
+				if (adapter->eee_advert &
+				    hw->dev_spec.ich8lan.eee_lp_ability &
+				    I82579_EEE_1000_SUPPORTED)
+					lpi_ctrl |= I82579_LPI_CTRL_1000_ENABLE;
+
+				retval = e1e_wphy_locked(hw, I82579_LPI_CTRL,
+							 lpi_ctrl);
+			}
+		}
+		hw->phy.ops.release(hw);
+	}
 
 	/* Release control of h/w to f/w.  If f/w is AMT enabled, this
 	 * would have already happened in close and is redundant.
@@ -6466,7 +6504,7 @@ static int __e1000_resume(struct pci_dev *pdev)
 	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L1)
 		aspm_disable_flag |= PCIE_LINK_STATE_L1;
 	if (aspm_disable_flag)
-		e1000e_disable_aspm_locked(pdev, aspm_disable_flag);
+		e1000e_disable_aspm(pdev, aspm_disable_flag);
 
 	pci_set_master(pdev);
 
@@ -6744,7 +6782,7 @@ static pci_ers_result_t e1000_io_slot_reset(struct pci_dev *pdev)
 	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L1)
 		aspm_disable_flag |= PCIE_LINK_STATE_L1;
 	if (aspm_disable_flag)
-		e1000e_disable_aspm(pdev, aspm_disable_flag);
+		e1000e_disable_aspm_locked(pdev, aspm_disable_flag);
 
 	err = pci_enable_device_mem(pdev);
 	if (err) {

drivers/net/ethernet/intel/e1000e/regs.h

@@ -125,7 +125,6 @@
 				 (0x054E4 + ((_i - 16) * 8)))
 #define E1000_SHRAL(_i)		(0x05438 + ((_i) * 8))
 #define E1000_SHRAH(_i)		(0x0543C + ((_i) * 8))
-#define E1000_TARC0_CB_MULTIQ_3_REQ	(1 << 28 | 1 << 29)
 #define E1000_TDFH		0x03410	/* Tx Data FIFO Head - RW */
 #define E1000_TDFT		0x03418	/* Tx Data FIFO Tail - RW */
 #define E1000_TDFHS		0x03420	/* Tx Data FIFO Head Saved - RW */

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

@@ -79,10 +79,13 @@
 #define I40E_MIN_MSIX                 2
 #define I40E_DEFAULT_NUM_VMDQ_VSI     8 /* max 256 VSIs */
 #define I40E_MIN_VSI_ALLOC            51 /* LAN, ATR, FCOE, 32 VF, 16 VMDQ */
-#define I40E_DEFAULT_QUEUES_PER_VMDQ  2 /* max 16 qps */
+/* max 16 qps */
+#define i40e_default_queues_per_vmdq(pf) \
+		(((pf)->flags & I40E_FLAG_RSS_AQ_CAPABLE) ? 4 : 1)
 #define I40E_DEFAULT_QUEUES_PER_VF    4
 #define I40E_DEFAULT_QUEUES_PER_TC    1 /* should be a power of 2 */
-#define I40E_MAX_QUEUES_PER_TC        64 /* should be a power of 2 */
+#define i40e_pf_get_max_q_per_tc(pf) \
+		(((pf)->flags & I40E_FLAG_128_QP_RSS_CAPABLE) ? 128 : 64)
 #define I40E_FDIR_RING                0
 #define I40E_FDIR_RING_COUNT          32
 #ifdef I40E_FCOE
@@ -298,6 +301,7 @@ struct i40e_pf {
 #define I40E_FLAG_VMDQ_ENABLED			BIT_ULL(7)
 #define I40E_FLAG_FDIR_REQUIRES_REINIT		BIT_ULL(8)
 #define I40E_FLAG_NEED_LINK_UPDATE		BIT_ULL(9)
+#define I40E_FLAG_IWARP_ENABLED			BIT_ULL(10)
 #ifdef I40E_FCOE
 #define I40E_FLAG_FCOE_ENABLED			BIT_ULL(11)
 #endif /* I40E_FCOE */
@@ -318,6 +322,12 @@ struct i40e_pf {
 #endif
 #define I40E_FLAG_PORT_ID_VALID			BIT_ULL(28)
 #define I40E_FLAG_DCB_CAPABLE			BIT_ULL(29)
+#define I40E_FLAG_RSS_AQ_CAPABLE		BIT_ULL(31)
+#define I40E_FLAG_HW_ATR_EVICT_CAPABLE		BIT_ULL(32)
+#define I40E_FLAG_OUTER_UDP_CSUM_CAPABLE	BIT_ULL(33)
+#define I40E_FLAG_128_QP_RSS_CAPABLE		BIT_ULL(34)
+#define I40E_FLAG_WB_ON_ITR_CAPABLE		BIT_ULL(35)
+#define I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE	BIT_ULL(38)
 #define I40E_FLAG_VEB_MODE_ENABLED		BIT_ULL(40)
 
 	/* tracks features that get auto disabled by errors */
@@ -550,6 +560,7 @@ struct i40e_q_vector {
 	cpumask_t affinity_mask;
 	struct rcu_head rcu;	/* to avoid race with update stats on free */
 	char name[I40E_INT_NAME_STR_LEN];
+	bool arm_wb_state;
 } ____cacheline_internodealigned_in_smp;
 
 /* lan device */

drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h

@@ -257,6 +257,10 @@ enum i40e_admin_queue_opc {
 	/* Tunnel commands */
 	i40e_aqc_opc_add_udp_tunnel	= 0x0B00,
 	i40e_aqc_opc_del_udp_tunnel	= 0x0B01,
+	i40e_aqc_opc_set_rss_key	= 0x0B02,
+	i40e_aqc_opc_set_rss_lut	= 0x0B03,
+	i40e_aqc_opc_get_rss_key	= 0x0B04,
+	i40e_aqc_opc_get_rss_lut	= 0x0B05,
 
 	/* Async Events */
 	i40e_aqc_opc_event_lan_overflow		= 0x1001,
@@ -821,8 +825,12 @@ struct i40e_aqc_vsi_properties_data {
 					 I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT)
 	/* queueing option section */
 	u8	queueing_opt_flags;
+#define I40E_AQ_VSI_QUE_OPT_MULTICAST_UDP_ENA	0x04
+#define I40E_AQ_VSI_QUE_OPT_UNICAST_UDP_ENA	0x08
 #define I40E_AQ_VSI_QUE_OPT_TCP_ENA	0x10
 #define I40E_AQ_VSI_QUE_OPT_FCOE_ENA	0x20
+#define I40E_AQ_VSI_QUE_OPT_RSS_LUT_PF	0x00
+#define I40E_AQ_VSI_QUE_OPT_RSS_LUT_VSI	0x40
 	u8	queueing_opt_reserved[3];
 	/* scheduler section */
 	u8	up_enable_bits;
@@ -2179,6 +2187,46 @@ struct i40e_aqc_del_udp_tunnel_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_del_udp_tunnel_completion);
 
+struct i40e_aqc_get_set_rss_key {
+#define I40E_AQC_SET_RSS_KEY_VSI_VALID		(0x1 << 15)
+#define I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT	0
+#define I40E_AQC_SET_RSS_KEY_VSI_ID_MASK	(0x3FF << \
+					I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT)
+	__le16	vsi_id;
+	u8	reserved[6];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_get_set_rss_key);
+
+struct i40e_aqc_get_set_rss_key_data {
+	u8 standard_rss_key[0x28];
+	u8 extended_hash_key[0xc];
+};
+
+I40E_CHECK_STRUCT_LEN(0x34, i40e_aqc_get_set_rss_key_data);
+
+struct  i40e_aqc_get_set_rss_lut {
+#define I40E_AQC_SET_RSS_LUT_VSI_VALID		(0x1 << 15)
+#define I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT	0
+#define I40E_AQC_SET_RSS_LUT_VSI_ID_MASK	(0x3FF << \
+					I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT)
+	__le16	vsi_id;
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT	0
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK	(0x1 << \
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT)
+
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI	0
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF	1
+	__le16	flags;
+	u8	reserved[4];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_get_set_rss_lut);
+
 /* tunnel key structure 0x0B10 */
 
 struct i40e_aqc_tunnel_key_structure {

drivers/net/ethernet/intel/i40e/i40e_common.c

@@ -54,6 +54,15 @@ static i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_20G_KR2:
 			hw->mac.type = I40E_MAC_XL710;
 			break;
+		case I40E_DEV_ID_SFP_X722:
+		case I40E_DEV_ID_1G_BASE_T_X722:
+		case I40E_DEV_ID_10G_BASE_T_X722:
+			hw->mac.type = I40E_MAC_X722;
+			break;
+		case I40E_DEV_ID_X722_VF:
+		case I40E_DEV_ID_X722_VF_HV:
+			hw->mac.type = I40E_MAC_X722_VF;
+			break;
 		case I40E_DEV_ID_VF:
 		case I40E_DEV_ID_VF_HV:
 			hw->mac.type = I40E_MAC_VF;
@@ -383,6 +392,169 @@ i40e_status i40e_aq_queue_shutdown(struct i40e_hw *hw,
 	return status;
 }
 
+/**
+ * i40e_aq_get_set_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ * @set: set true to set the table, false to get the table
+ *
+ * Internal function to get or set RSS look up table
+ **/
+static i40e_status i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
+					   u16 vsi_id, bool pf_lut,
+					   u8 *lut, u16 lut_size,
+					   bool set)
+{
+	i40e_status status;
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_get_set_rss_lut *cmd_resp =
+		   (struct i40e_aqc_get_set_rss_lut *)&desc.params.raw;
+
+	if (set)
+		i40e_fill_default_direct_cmd_desc(&desc,
+						  i40e_aqc_opc_set_rss_lut);
+	else
+		i40e_fill_default_direct_cmd_desc(&desc,
+						  i40e_aqc_opc_get_rss_lut);
+
+	/* Indirect command */
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+
+	cmd_resp->vsi_id =
+			cpu_to_le16((u16)((vsi_id <<
+					  I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT) &
+					  I40E_AQC_SET_RSS_LUT_VSI_ID_MASK));
+	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_LUT_VSI_VALID);
+
+	if (pf_lut)
+		cmd_resp->flags |= cpu_to_le16((u16)
+					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF <<
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+	else
+		cmd_resp->flags |= cpu_to_le16((u16)
+					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI <<
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+
+	cmd_resp->addr_high = cpu_to_le32(high_16_bits((u64)lut));
+	cmd_resp->addr_low = cpu_to_le32(lower_32_bits((u64)lut));
+
+	status = i40e_asq_send_command(hw, &desc, lut, lut_size, NULL);
+
+	return status;
+}
+
+/**
+ * i40e_aq_get_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ *
+ * get the RSS lookup table, PF or VSI type
+ **/
+i40e_status i40e_aq_get_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+				bool pf_lut, u8 *lut, u16 lut_size)
+{
+	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size,
+				       false);
+}
+
+/**
+ * i40e_aq_set_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ *
+ * set the RSS lookup table, PF or VSI type
+ **/
+i40e_status i40e_aq_set_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+				bool pf_lut, u8 *lut, u16 lut_size)
+{
+	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size, true);
+}
+
+/**
+ * i40e_aq_get_set_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ * @set: set true to set the key, false to get the key
+ *
+ * get the RSS key per VSI
+ **/
+static i40e_status i40e_aq_get_set_rss_key(struct i40e_hw *hw,
+				      u16 vsi_id,
+				      struct i40e_aqc_get_set_rss_key_data *key,
+				      bool set)
+{
+	i40e_status status;
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_get_set_rss_key *cmd_resp =
+			(struct i40e_aqc_get_set_rss_key *)&desc.params.raw;
+	u16 key_size = sizeof(struct i40e_aqc_get_set_rss_key_data);
+
+	if (set)
+		i40e_fill_default_direct_cmd_desc(&desc,
+						  i40e_aqc_opc_set_rss_key);
+	else
+		i40e_fill_default_direct_cmd_desc(&desc,
+						  i40e_aqc_opc_get_rss_key);
+
+	/* Indirect command */
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+
+	cmd_resp->vsi_id =
+			cpu_to_le16((u16)((vsi_id <<
+					  I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT) &
+					  I40E_AQC_SET_RSS_KEY_VSI_ID_MASK));
+	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_KEY_VSI_VALID);
+	cmd_resp->addr_high = cpu_to_le32(high_16_bits((u64)key));
+	cmd_resp->addr_low = cpu_to_le32(lower_32_bits((u64)key));
+
+	status = i40e_asq_send_command(hw, &desc, key, key_size, NULL);
+
+	return status;
+}
+
+/**
+ * i40e_aq_get_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ *
+ **/
+i40e_status i40e_aq_get_rss_key(struct i40e_hw *hw,
+				u16 vsi_id,
+				struct i40e_aqc_get_set_rss_key_data *key)
+{
+	return i40e_aq_get_set_rss_key(hw, vsi_id, key, false);
+}
+
+/**
+ * i40e_aq_set_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ *
+ * set the RSS key per VSI
+ **/
+i40e_status i40e_aq_set_rss_key(struct i40e_hw *hw,
+				u16 vsi_id,
+				struct i40e_aqc_get_set_rss_key_data *key)
+{
+	return i40e_aq_get_set_rss_key(hw, vsi_id, key, true);
+}
+
 /* The i40e_ptype_lookup table is used to convert from the 8-bit ptype in the
  * hardware to a bit-field that can be used by SW to more easily determine the
  * packet type.
@@ -769,6 +941,7 @@ i40e_status i40e_init_shared_code(struct i40e_hw *hw)
 
 	switch (hw->mac.type) {
 	case I40E_MAC_XL710:
+	case I40E_MAC_X722:
 		break;
 	default:
 		return I40E_ERR_DEVICE_NOT_SUPPORTED;

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

@@ -76,6 +76,9 @@ static const struct pci_device_id i40e_pci_tbl[] = {
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_C), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T), 0},
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0},
 	/* required last entry */
 	{0, }
 };
@@ -1547,7 +1550,7 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
 	 */
 	qcount = min_t(int, vsi->alloc_queue_pairs, pf->num_lan_msix);
 	num_tc_qps = qcount / numtc;
-	num_tc_qps = min_t(int, num_tc_qps, I40E_MAX_QUEUES_PER_TC);
+	num_tc_qps = min_t(int, num_tc_qps, i40e_pf_get_max_q_per_tc(pf));
 
 	/* Setup queue offset/count for all TCs for given VSI */
 	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
@@ -2905,6 +2908,9 @@ static void i40e_enable_misc_int_causes(struct i40e_pf *pf)
 	      I40E_PFINT_ICR0_ENA_VFLR_MASK          |
 	      I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
 
+	if (pf->flags & I40E_FLAG_IWARP_ENABLED)
+		val |= I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK;
+
 	if (pf->flags & I40E_FLAG_PTP)
 		val |= I40E_PFINT_ICR0_ENA_TIMESYNC_MASK;
 
@@ -3195,6 +3201,13 @@ static irqreturn_t i40e_intr(int irq, void *data)
 	    (icr0 & I40E_PFINT_ICR0_SWINT_MASK))
 		pf->sw_int_count++;
 
+	if ((pf->flags & I40E_FLAG_IWARP_ENABLED) &&
+	    (ena_mask & I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK)) {
+		ena_mask &= ~I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK;
+		icr0 &= ~I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK;
+		dev_info(&pf->pdev->dev, "cleared PE_CRITERR\n");
+	}
+
 	/* only q0 is used in MSI/Legacy mode, and none are used in MSIX */
 	if (icr0 & I40E_PFINT_ICR0_QUEUE_0_MASK) {
 
@@ -7068,6 +7081,10 @@ static int i40e_alloc_rings(struct i40e_vsi *vsi)
 		tx_ring->count = vsi->num_desc;
 		tx_ring->size = 0;
 		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];
@@ -7466,62 +7483,139 @@ static int i40e_setup_misc_vector(struct i40e_pf *pf)
 }
 
 /**
- * i40e_config_rss - Prepare for RSS if used
+ * i40e_config_rss_aq - Prepare for RSS using AQ commands
+ * @vsi: vsi structure
+ * @seed: RSS hash seed
+ **/
+static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed)
+{
+	struct i40e_aqc_get_set_rss_key_data rss_key;
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
+	bool pf_lut = false;
+	u8 *rss_lut;
+	int ret, i;
+
+	memset(&rss_key, 0, sizeof(rss_key));
+	memcpy(&rss_key, seed, sizeof(rss_key));
+
+	rss_lut = kzalloc(pf->rss_table_size, GFP_KERNEL);
+	if (!rss_lut)
+		return -ENOMEM;
+
+	/* Populate the LUT with max no. of queues in round robin fashion */
+	for (i = 0; i < vsi->rss_table_size; i++)
+		rss_lut[i] = i % vsi->rss_size;
+
+	ret = i40e_aq_set_rss_key(hw, vsi->id, &rss_key);
+	if (ret) {
+		dev_info(&pf->pdev->dev,
+			 "Cannot set RSS key, err %s aq_err %s\n",
+			 i40e_stat_str(&pf->hw, ret),
+			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
+		return ret;
+	}
+
+	if (vsi->type == I40E_VSI_MAIN)
+		pf_lut = true;
+
+	ret = i40e_aq_set_rss_lut(hw, vsi->id, pf_lut, rss_lut,
+				  vsi->rss_table_size);
+	if (ret)
+		dev_info(&pf->pdev->dev,
+			 "Cannot set RSS lut, err %s aq_err %s\n",
+			 i40e_stat_str(&pf->hw, ret),
+			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
+
+	return ret;
+}
+
+/**
+ * i40e_vsi_config_rss - Prepare for VSI(VMDq) RSS if used
+ * @vsi: VSI structure
+ **/
+static int i40e_vsi_config_rss(struct i40e_vsi *vsi)
+{
+	u8 seed[I40E_HKEY_ARRAY_SIZE];
+	struct i40e_pf *pf = vsi->back;
+
+	netdev_rss_key_fill((void *)seed, I40E_HKEY_ARRAY_SIZE);
+	vsi->rss_size = min_t(int, pf->rss_size, vsi->num_queue_pairs);
+
+	if (pf->flags & I40E_FLAG_RSS_AQ_CAPABLE)
+		return i40e_config_rss_aq(vsi, seed);
+
+	return 0;
+}
+
+/**
+ * i40e_config_rss_reg - Prepare for RSS if used
  * @pf: board private structure
+ * @seed: RSS hash seed
  **/
-static int i40e_config_rss(struct i40e_pf *pf)
+static int i40e_config_rss_reg(struct i40e_pf *pf, const u8 *seed)
 {
-	u32 rss_key[I40E_PFQF_HKEY_MAX_INDEX + 1];
 	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
 	struct i40e_hw *hw = &pf->hw;
+	u32 *seed_dw = (u32 *)seed;
+	u32 current_queue = 0;
 	u32 lut = 0;
 	int i, j;
-	u64 hena;
-	u32 reg_val;
 
-	netdev_rss_key_fill(rss_key, sizeof(rss_key));
+	/* Fill out hash function seed */
 	for (i = 0; i <= I40E_PFQF_HKEY_MAX_INDEX; i++)
-		wr32(hw, I40E_PFQF_HKEY(i), rss_key[i]);
+		wr32(hw, I40E_PFQF_HKEY(i), seed_dw[i]);
+
+	for (i = 0; i <= I40E_PFQF_HLUT_MAX_INDEX; i++) {
+		lut = 0;
+		for (j = 0; j < 4; j++) {
+			if (current_queue == vsi->rss_size)
+				current_queue = 0;
+			lut |= ((current_queue) << (8 * j));
+			current_queue++;
+		}
+		wr32(&pf->hw, I40E_PFQF_HLUT(i), lut);
+	}
+	i40e_flush(hw);
+
+	return 0;
+}
+
+/**
+ * i40e_config_rss - Prepare for RSS if used
+ * @pf: board private structure
+ **/
+static int i40e_config_rss(struct i40e_pf *pf)
+{
+	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
+	u8 seed[I40E_HKEY_ARRAY_SIZE];
+	struct i40e_hw *hw = &pf->hw;
+	u32 reg_val;
+	u64 hena;
+
+	netdev_rss_key_fill((void *)seed, I40E_HKEY_ARRAY_SIZE);
 
 	/* By default we enable TCP/UDP with IPv4/IPv6 ptypes */
 	hena = (u64)rd32(hw, I40E_PFQF_HENA(0)) |
 		((u64)rd32(hw, I40E_PFQF_HENA(1)) << 32);
-	hena |= I40E_DEFAULT_RSS_HENA;
+	hena |= i40e_pf_get_default_rss_hena(pf);
+
 	wr32(hw, I40E_PFQF_HENA(0), (u32)hena);
 	wr32(hw, I40E_PFQF_HENA(1), (u32)(hena >> 32));
 
 	vsi->rss_size = min_t(int, pf->rss_size, vsi->num_queue_pairs);
 
-	/* Check capability and Set table size and register per hw expectation*/
+	/* Determine the RSS table size based on the hardware capabilities */
 	reg_val = rd32(hw, I40E_PFQF_CTL_0);
-	if (pf->rss_table_size == 512)
-		reg_val |= I40E_PFQF_CTL_0_HASHLUTSIZE_512;
-	else
-		reg_val &= ~I40E_PFQF_CTL_0_HASHLUTSIZE_512;
+	reg_val = (pf->rss_table_size == 512) ?
+			(reg_val | I40E_PFQF_CTL_0_HASHLUTSIZE_512) :
+			(reg_val & ~I40E_PFQF_CTL_0_HASHLUTSIZE_512);
 	wr32(hw, I40E_PFQF_CTL_0, reg_val);
 
-	/* Populate the LUT with max no. of queues in round robin fashion */
-	for (i = 0, j = 0; i < pf->rss_table_size; i++, j++) {
-
-		/* The assumption is that lan qp count will be the highest
-		 * qp count for any PF VSI that needs RSS.
-		 * If multiple VSIs need RSS support, all the qp counts
-		 * for those VSIs should be a power of 2 for RSS to work.
-		 * If LAN VSI is the only consumer for RSS then this requirement
-		 * is not necessary.
-		 */
-		if (j == vsi->rss_size)
-			j = 0;
-		/* lut = 4-byte sliding window of 4 lut entries */
-		lut = (lut << 8) | (j &
-			 (BIT(pf->hw.func_caps.rss_table_entry_width) - 1));
-		/* On i = 3, we have 4 entries in lut; write to the register */
-		if ((i & 3) == 3)
-			wr32(hw, I40E_PFQF_HLUT(i >> 2), lut);
-	}
-	i40e_flush(hw);
-
-	return 0;
+	if (pf->flags & I40E_FLAG_RSS_AQ_CAPABLE)
+		return i40e_config_rss_aq(pf->vsi[pf->lan_vsi], seed);
+	else
+		return i40e_config_rss_reg(pf, seed);
 }
 
 /**
@@ -7762,9 +7856,8 @@ static int i40e_sw_init(struct i40e_pf *pf)
 	}
 
 	if (pf->hw.func_caps.vmdq) {
-		pf->flags |= I40E_FLAG_VMDQ_ENABLED;
 		pf->num_vmdq_vsis = I40E_DEFAULT_NUM_VMDQ_VSI;
-		pf->num_vmdq_qps = I40E_DEFAULT_QUEUES_PER_VMDQ;
+		pf->flags |= I40E_FLAG_VMDQ_ENABLED;
 	}
 
 #ifdef I40E_FCOE
@@ -7782,6 +7875,14 @@ static int i40e_sw_init(struct i40e_pf *pf)
 					I40E_MAX_VF_COUNT);
 	}
 #endif /* CONFIG_PCI_IOV */
+	if (pf->hw.mac.type == I40E_MAC_X722) {
+		pf->flags |= I40E_FLAG_RSS_AQ_CAPABLE |
+			     I40E_FLAG_128_QP_RSS_CAPABLE |
+			     I40E_FLAG_HW_ATR_EVICT_CAPABLE |
+			     I40E_FLAG_OUTER_UDP_CSUM_CAPABLE |
+			     I40E_FLAG_WB_ON_ITR_CAPABLE |
+			     I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE;
+	}
 	pf->eeprom_version = 0xDEAD;
 	pf->lan_veb = I40E_NO_VEB;
 	pf->lan_vsi = I40E_NO_VSI;
@@ -8937,6 +9038,10 @@ struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf, u8 type,
 		break;
 	}
 
+	if ((pf->flags & I40E_FLAG_RSS_AQ_CAPABLE) &&
+	    (vsi->type == I40E_VSI_VMDQ2)) {
+		ret = i40e_vsi_config_rss(vsi);
+	}
 	return vsi;
 
 err_rings:

drivers/net/ethernet/intel/i40e/i40e_nvm.c

@@ -211,6 +211,74 @@ static i40e_status i40e_read_nvm_word_srctl(struct i40e_hw *hw, u16 offset,
 	return ret_code;
 }
 
+/**
+ * i40e_read_nvm_aq - Read Shadow RAM.
+ * @hw: pointer to the HW structure.
+ * @module_pointer: module pointer location in words from the NVM beginning
+ * @offset: offset in words from module start
+ * @words: number of words to write
+ * @data: buffer with words to write to the Shadow RAM
+ * @last_command: tells the AdminQ that this is the last command
+ *
+ * Writes a 16 bit words buffer to the Shadow RAM using the admin command.
+ **/
+static i40e_status i40e_read_nvm_aq(struct i40e_hw *hw, u8 module_pointer,
+				    u32 offset, u16 words, void *data,
+				    bool last_command)
+{
+	i40e_status ret_code = I40E_ERR_NVM;
+	struct i40e_asq_cmd_details cmd_details;
+
+	memset(&cmd_details, 0, sizeof(cmd_details));
+
+	/* Here we are checking the SR limit only for the flat memory model.
+	 * We cannot do it for the module-based model, as we did not acquire
+	 * the NVM resource yet (we cannot get the module pointer value).
+	 * Firmware will check the module-based model.
+	 */
+	if ((offset + words) > hw->nvm.sr_size)
+		i40e_debug(hw, I40E_DEBUG_NVM,
+			   "NVM write error: offset %d beyond Shadow RAM limit %d\n",
+			   (offset + words), hw->nvm.sr_size);
+	else if (words > I40E_SR_SECTOR_SIZE_IN_WORDS)
+		/* We can write only up to 4KB (one sector), in one AQ write */
+		i40e_debug(hw, I40E_DEBUG_NVM,
+			   "NVM write fail error: tried to write %d words, limit is %d.\n",
+			   words, I40E_SR_SECTOR_SIZE_IN_WORDS);
+	else if (((offset + (words - 1)) / I40E_SR_SECTOR_SIZE_IN_WORDS)
+		 != (offset / I40E_SR_SECTOR_SIZE_IN_WORDS))
+		/* A single write cannot spread over two sectors */
+		i40e_debug(hw, I40E_DEBUG_NVM,
+			   "NVM write error: cannot spread over two sectors in a single write offset=%d words=%d\n",
+			   offset, words);
+	else
+		ret_code = i40e_aq_read_nvm(hw, module_pointer,
+					    2 * offset,  /*bytes*/
+					    2 * words,   /*bytes*/
+					    data, last_command, &cmd_details);
+
+	return ret_code;
+}
+
+/**
+ * i40e_read_nvm_word_aq - Reads Shadow RAM via AQ
+ * @hw: pointer to the HW structure
+ * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF)
+ * @data: word read from the Shadow RAM
+ *
+ * Reads one 16 bit word from the Shadow RAM using the GLNVM_SRCTL register.
+ **/
+static i40e_status i40e_read_nvm_word_aq(struct i40e_hw *hw, u16 offset,
+					 u16 *data)
+{
+	i40e_status ret_code = I40E_ERR_TIMEOUT;
+
+	ret_code = i40e_read_nvm_aq(hw, 0x0, offset, 1, data, true);
+	*data = le16_to_cpu(*(__le16 *)data);
+
+	return ret_code;
+}
+
 /**
  * i40e_read_nvm_word - Reads Shadow RAM
  * @hw: pointer to the HW structure
@@ -222,6 +290,8 @@ static i40e_status i40e_read_nvm_word_srctl(struct i40e_hw *hw, u16 offset,
 i40e_status i40e_read_nvm_word(struct i40e_hw *hw, u16 offset,
 			       u16 *data)
 {
+	if (hw->mac.type == I40E_MAC_X722)
+		return i40e_read_nvm_word_aq(hw, offset, data);
 	return i40e_read_nvm_word_srctl(hw, offset, data);
 }
 
@@ -256,6 +326,63 @@ static i40e_status i40e_read_nvm_buffer_srctl(struct i40e_hw *hw, u16 offset,
 	return ret_code;
 }
 
+/**
+ * i40e_read_nvm_buffer_aq - Reads Shadow RAM buffer via AQ
+ * @hw: pointer to the HW structure
+ * @offset: offset of the Shadow RAM word to read (0x000000 - 0x001FFF).
+ * @words: (in) number of words to read; (out) number of words actually read
+ * @data: words read from the Shadow RAM
+ *
+ * Reads 16 bit words (data buffer) from the SR using the i40e_read_nvm_aq()
+ * method. The buffer read is preceded by the NVM ownership take
+ * and followed by the release.
+ **/
+static i40e_status i40e_read_nvm_buffer_aq(struct i40e_hw *hw, u16 offset,
+					   u16 *words, u16 *data)
+{
+	i40e_status ret_code;
+	u16 read_size = *words;
+	bool last_cmd = false;
+	u16 words_read = 0;
+	u16 i = 0;
+
+	do {
+		/* Calculate number of bytes we should read in this step.
+		 * FVL AQ do not allow to read more than one page at a time or
+		 * to cross page boundaries.
+		 */
+		if (offset % I40E_SR_SECTOR_SIZE_IN_WORDS)
+			read_size = min(*words,
+					(u16)(I40E_SR_SECTOR_SIZE_IN_WORDS -
+				      (offset % I40E_SR_SECTOR_SIZE_IN_WORDS)));
+		else
+			read_size = min((*words - words_read),
+					I40E_SR_SECTOR_SIZE_IN_WORDS);
+
+		/* Check if this is last command, if so set proper flag */
+		if ((words_read + read_size) >= *words)
+			last_cmd = true;
+
+		ret_code = i40e_read_nvm_aq(hw, 0x0, offset, read_size,
+					    data + words_read, last_cmd);
+		if (ret_code)
+			goto read_nvm_buffer_aq_exit;
+
+		/* Increment counter for words already read and move offset to
+		 * new read location
+		 */
+		words_read += read_size;
+		offset += read_size;
+	} while (words_read < *words);
+
+	for (i = 0; i < *words; i++)
+		data[i] = le16_to_cpu(((__le16 *)data)[i]);
+
+read_nvm_buffer_aq_exit:
+	*words = words_read;
+	return ret_code;
+}
+
 /**
  * i40e_read_nvm_buffer - Reads Shadow RAM buffer
  * @hw: pointer to the HW structure
@@ -270,6 +397,8 @@ static i40e_status i40e_read_nvm_buffer_srctl(struct i40e_hw *hw, u16 offset,
 i40e_status i40e_read_nvm_buffer(struct i40e_hw *hw, u16 offset,
 				 u16 *words, u16 *data)
 {
+	if (hw->mac.type == I40E_MAC_X722)
+		return i40e_read_nvm_buffer_aq(hw, offset, words, data);
 	return i40e_read_nvm_buffer_srctl(hw, offset, words, data);
 }
 

drivers/net/ethernet/intel/i40e/i40e_prototype.h

@@ -61,6 +61,17 @@ i40e_status i40e_aq_queue_shutdown(struct i40e_hw *hw, bool unloading);
 char *i40e_aq_str(struct i40e_hw *hw, enum i40e_admin_queue_err aq_err);
 char *i40e_stat_str(struct i40e_hw *hw, i40e_status stat_err);
 
+i40e_status i40e_aq_get_rss_lut(struct i40e_hw *hw, u16 seid,
+				bool pf_lut, u8 *lut, u16 lut_size);
+i40e_status i40e_aq_set_rss_lut(struct i40e_hw *hw, u16 seid,
+				bool pf_lut, u8 *lut, u16 lut_size);
+i40e_status i40e_aq_get_rss_key(struct i40e_hw *hw,
+				u16 seid,
+				struct i40e_aqc_get_set_rss_key_data *key);
+i40e_status i40e_aq_set_rss_key(struct i40e_hw *hw,
+				u16 seid,
+				struct i40e_aqc_get_set_rss_key_data *key);
+
 u32 i40e_led_get(struct i40e_hw *hw);
 void i40e_led_set(struct i40e_hw *hw, u32 mode, bool blink);
 

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

@@ -853,6 +853,22 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
  **/
 static void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
 {
+	u16 flags = q_vector->tx.ring[0].flags;
+
+	if (flags & I40E_TXR_FLAGS_WB_ON_ITR) {
+		u32 val;
+
+		if (q_vector->arm_wb_state)
+			return;
+
+		val = I40E_PFINT_DYN_CTLN_WB_ON_ITR_MASK;
+
+		wr32(&vsi->back->hw,
+		     I40E_PFINT_DYN_CTLN(q_vector->v_idx +
+					 vsi->base_vector - 1),
+		     val);
+		q_vector->arm_wb_state = true;
+	} else if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
 		u32 val = I40E_PFINT_DYN_CTLN_INTENA_MASK |
 			  I40E_PFINT_DYN_CTLN_ITR_INDX_MASK | /* set noitr */
 			  I40E_PFINT_DYN_CTLN_SWINT_TRIG_MASK |
@@ -860,8 +876,17 @@ static void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
 			  /* allow 00 to be written to the index */
 
 		wr32(&vsi->back->hw,
-	     I40E_PFINT_DYN_CTLN(q_vector->v_idx + vsi->base_vector - 1),
-	     val);
+		     I40E_PFINT_DYN_CTLN(q_vector->v_idx +
+					 vsi->base_vector - 1), val);
+	} else {
+		u32 val = I40E_PFINT_DYN_CTL0_INTENA_MASK |
+			  I40E_PFINT_DYN_CTL0_ITR_INDX_MASK | /* set noitr */
+			  I40E_PFINT_DYN_CTL0_SWINT_TRIG_MASK |
+			  I40E_PFINT_DYN_CTL0_SW_ITR_INDX_ENA_MASK;
+			/* allow 00 to be written to the index */
+
+		wr32(&vsi->back->hw, I40E_PFINT_DYN_CTL0, val);
+	}
 }
 
 /**
@@ -1404,7 +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);
@@ -1918,6 +1944,9 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 		return budget;
 	}
 
+	if (vsi->back->flags & I40E_TXR_FLAGS_WB_ON_ITR)
+		q_vector->arm_wb_state = false;
+
 	/* Work is done so exit the polling mode and re-enable the interrupt */
 	napi_complete(napi);
 	if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED) {
@@ -2011,6 +2040,13 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	/* Due to lack of space, no more new filters can be programmed */
 	if (th->syn && (pf->auto_disable_flags & I40E_FLAG_FD_ATR_ENABLED))
 		return;
+	if (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE) {
+		/* HW ATR eviction will take care of removing filters on FIN
+		 * and RST packets.
+		 */
+		if (th->fin || th->rst)
+			return;
+	}
 
 	tx_ring->atr_count++;
 
@@ -2066,6 +2102,9 @@ static void i40e_atr(struct i40e_ring *tx_ring, struct sk_buff *skb,
 			I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT) &
 			I40E_TXD_FLTR_QW1_CNTINDEX_MASK;
 
+	if (pf->flags & I40E_FLAG_HW_ATR_EVICT_CAPABLE)
+		dtype_cmd |= I40E_TXD_FLTR_QW1_ATR_MASK;
+
 	fdir_desc->qindex_flex_ptype_vsi = cpu_to_le32(flex_ptype);
 	fdir_desc->rsvd = cpu_to_le32(0);
 	fdir_desc->dtype_cmd_cntindex = cpu_to_le32(dtype_cmd);
@@ -2273,11 +2312,15 @@ 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;
@@ -2314,6 +2357,15 @@ 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

@@ -78,6 +78,18 @@ enum i40e_dyn_idx_t {
 	BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6) | \
 	BIT_ULL(I40E_FILTER_PCTYPE_L2_PAYLOAD))
 
+#define I40E_DEFAULT_RSS_HENA_EXPANDED (I40E_DEFAULT_RSS_HENA | \
+	BIT(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK) | \
+	BIT(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
+	BIT(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
+	BIT(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK) | \
+	BIT(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | \
+	BIT(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP))
+
+#define i40e_pf_get_default_rss_hena(pf) \
+	(((pf)->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE) ? \
+	  I40E_DEFAULT_RSS_HENA_EXPANDED : I40E_DEFAULT_RSS_HENA)
+
 /* Supported Rx Buffer Sizes */
 #define I40E_RXBUFFER_512   512    /* Used for packet split */
 #define I40E_RXBUFFER_2048  2048
@@ -253,6 +265,10 @@ struct i40e_ring {
 	bool ring_active;		/* is ring online or not */
 	bool arm_wb;		/* do something to arm write back */
 
+	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

@@ -47,6 +47,11 @@
 #define I40E_DEV_ID_20G_KR2		0x1587
 #define I40E_DEV_ID_VF			0x154C
 #define I40E_DEV_ID_VF_HV		0x1571
+#define I40E_DEV_ID_SFP_X722		0x37D0
+#define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
+#define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
+#define I40E_DEV_ID_X722_VF		0x37CD
+#define I40E_DEV_ID_X722_VF_HV		0x37D9
 
 #define i40e_is_40G_device(d)		((d) == I40E_DEV_ID_QSFP_A  || \
 					 (d) == I40E_DEV_ID_QSFP_B  || \
@@ -120,6 +125,8 @@ enum i40e_mac_type {
 	I40E_MAC_X710,
 	I40E_MAC_XL710,
 	I40E_MAC_VF,
+	I40E_MAC_X722,
+	I40E_MAC_X722_VF,
 	I40E_MAC_GENERIC,
 };
 
@@ -502,7 +509,8 @@ struct i40e_hw {
 
 static inline bool i40e_is_vf(struct i40e_hw *hw)
 {
-	return hw->mac.type == I40E_MAC_VF;
+	return (hw->mac.type == I40E_MAC_VF ||
+		hw->mac.type == I40E_MAC_X722_VF);
 }
 
 struct i40e_driver_version {
@@ -599,14 +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!!! */
 };
 
@@ -947,6 +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;
@@ -965,15 +979,24 @@ struct i40e_filter_program_desc {
 
 /* Packet Classifier Types for filters */
 enum i40e_filter_pctype {
-	/* Note: Values 0-30 are reserved for future use */
+	/* Note: Values 0-28 are reserved for future use.
+	 * Value 29, 30, 32 are not supported on XL710 and X710.
+	 */
+	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP	= 29,
+	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP	= 30,
 	I40E_FILTER_PCTYPE_NONF_IPV4_UDP		= 31,
-	/* Note: Value 32 is reserved for future use */
+	I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK	= 32,
 	I40E_FILTER_PCTYPE_NONF_IPV4_TCP		= 33,
 	I40E_FILTER_PCTYPE_NONF_IPV4_SCTP		= 34,
 	I40E_FILTER_PCTYPE_NONF_IPV4_OTHER		= 35,
 	I40E_FILTER_PCTYPE_FRAG_IPV4			= 36,
-	/* Note: Values 37-40 are reserved for future use */
+	/* Note: Values 37-38 are reserved for future use.
+	 * Value 39, 40, 42 are not supported on XL710 and X710.
+	 */
+	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP	= 39,
+	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP	= 40,
 	I40E_FILTER_PCTYPE_NONF_IPV6_UDP		= 41,
+	I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK	= 42,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP		= 43,
 	I40E_FILTER_PCTYPE_NONF_IPV6_SCTP		= 44,
 	I40E_FILTER_PCTYPE_NONF_IPV6_OTHER		= 45,
@@ -1026,6 +1049,10 @@ enum i40e_filter_program_desc_pcmd {
 #define I40E_TXD_FLTR_QW1_FD_STATUS_MASK (0x3ULL << \
 					  I40E_TXD_FLTR_QW1_FD_STATUS_SHIFT)
 
+#define I40E_TXD_FLTR_QW1_ATR_SHIFT	(0xEULL + \
+					 I40E_TXD_FLTR_QW1_CMD_SHIFT)
+#define I40E_TXD_FLTR_QW1_ATR_MASK	BIT_ULL(I40E_TXD_FLTR_QW1_ATR_SHIFT)
+
 #define I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT 20
 #define I40E_TXD_FLTR_QW1_CNTINDEX_MASK	(0x1FFUL << \
 					 I40E_TXD_FLTR_QW1_CNTINDEX_SHIFT)

drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c

@@ -1177,9 +1177,14 @@ static int i40e_vc_get_vf_resources_msg(struct i40e_vf *vf, u8 *msg)
 	vfres->vf_offload_flags = I40E_VIRTCHNL_VF_OFFLOAD_L2;
 	vsi = pf->vsi[vf->lan_vsi_idx];
 	if (!vsi->info.pvid)
-		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_VLAN |
-					   I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG;
-
+		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_VLAN;
+	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;
+	} else {
+		vfres->vf_offload_flags |= I40E_VIRTCHNL_VF_OFFLOAD_RSS_REG;
+	}
 	vfres->num_vsis = num_vsis;
 	vfres->num_queue_pairs = vf->num_queue_pairs;
 	vfres->max_vectors = pf->hw.func_caps.num_msix_vectors_vf;

drivers/net/ethernet/intel/i40evf/i40e_adminq_cmd.h

@@ -35,7 +35,6 @@
 
 #define I40E_FW_API_VERSION_MAJOR	0x0001
 #define I40E_FW_API_VERSION_MINOR	0x0004
-#define I40E_FW_API_VERSION_A0_MINOR  0x0000
 
 struct i40e_aq_desc {
 	__le16 flags;
@@ -255,6 +254,10 @@ enum i40e_admin_queue_opc {
 	/* Tunnel commands */
 	i40e_aqc_opc_add_udp_tunnel	= 0x0B00,
 	i40e_aqc_opc_del_udp_tunnel	= 0x0B01,
+	i40e_aqc_opc_set_rss_key	= 0x0B02,
+	i40e_aqc_opc_set_rss_lut	= 0x0B03,
+	i40e_aqc_opc_get_rss_key	= 0x0B04,
+	i40e_aqc_opc_get_rss_lut	= 0x0B05,
 
 	/* Async Events */
 	i40e_aqc_opc_event_lan_overflow		= 0x1001,
@@ -819,8 +822,12 @@ struct i40e_aqc_vsi_properties_data {
 					 I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT)
 	/* queueing option section */
 	u8	queueing_opt_flags;
+#define I40E_AQ_VSI_QUE_OPT_MULTICAST_UDP_ENA	0x04
+#define I40E_AQ_VSI_QUE_OPT_UNICAST_UDP_ENA	0x08
 #define I40E_AQ_VSI_QUE_OPT_TCP_ENA	0x10
 #define I40E_AQ_VSI_QUE_OPT_FCOE_ENA	0x20
+#define I40E_AQ_VSI_QUE_OPT_RSS_LUT_PF	0x00
+#define I40E_AQ_VSI_QUE_OPT_RSS_LUT_VSI	0x40
 	u8	queueing_opt_reserved[3];
 	/* scheduler section */
 	u8	up_enable_bits;
@@ -2089,6 +2096,46 @@ struct i40e_aqc_del_udp_tunnel_completion {
 
 I40E_CHECK_CMD_LENGTH(i40e_aqc_del_udp_tunnel_completion);
 
+struct i40e_aqc_get_set_rss_key {
+#define I40E_AQC_SET_RSS_KEY_VSI_VALID		(0x1 << 15)
+#define I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT	0
+#define I40E_AQC_SET_RSS_KEY_VSI_ID_MASK	(0x3FF << \
+					I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT)
+	__le16	vsi_id;
+	u8	reserved[6];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_get_set_rss_key);
+
+struct i40e_aqc_get_set_rss_key_data {
+	u8 standard_rss_key[0x28];
+	u8 extended_hash_key[0xc];
+};
+
+I40E_CHECK_STRUCT_LEN(0x34, i40e_aqc_get_set_rss_key_data);
+
+struct  i40e_aqc_get_set_rss_lut {
+#define I40E_AQC_SET_RSS_LUT_VSI_VALID		(0x1 << 15)
+#define I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT	0
+#define I40E_AQC_SET_RSS_LUT_VSI_ID_MASK	(0x3FF << \
+					I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT)
+	__le16	vsi_id;
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT	0
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK	(0x1 << \
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT)
+
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI	0
+#define I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF	1
+	__le16	flags;
+	u8	reserved[4];
+	__le32	addr_high;
+	__le32	addr_low;
+};
+
+I40E_CHECK_CMD_LENGTH(i40e_aqc_get_set_rss_lut);
+
 /* tunnel key structure 0x0B10 */
 
 struct i40e_aqc_tunnel_key_structure_A0 {

drivers/net/ethernet/intel/i40evf/i40e_common.c

@@ -54,6 +54,15 @@ i40e_status i40e_set_mac_type(struct i40e_hw *hw)
 		case I40E_DEV_ID_20G_KR2:
 			hw->mac.type = I40E_MAC_XL710;
 			break;
+		case I40E_DEV_ID_SFP_X722:
+		case I40E_DEV_ID_1G_BASE_T_X722:
+		case I40E_DEV_ID_10G_BASE_T_X722:
+			hw->mac.type = I40E_MAC_X722;
+			break;
+		case I40E_DEV_ID_X722_VF:
+		case I40E_DEV_ID_X722_VF_HV:
+			hw->mac.type = I40E_MAC_X722_VF;
+			break;
 		case I40E_DEV_ID_VF:
 		case I40E_DEV_ID_VF_HV:
 			hw->mac.type = I40E_MAC_VF;
@@ -383,6 +392,169 @@ i40e_status i40evf_aq_queue_shutdown(struct i40e_hw *hw,
 	return status;
 }
 
+/**
+ * i40e_aq_get_set_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ * @set: set true to set the table, false to get the table
+ *
+ * Internal function to get or set RSS look up table
+ **/
+static i40e_status i40e_aq_get_set_rss_lut(struct i40e_hw *hw,
+					   u16 vsi_id, bool pf_lut,
+					   u8 *lut, u16 lut_size,
+					   bool set)
+{
+	i40e_status status;
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_get_set_rss_lut *cmd_resp =
+		   (struct i40e_aqc_get_set_rss_lut *)&desc.params.raw;
+
+	if (set)
+		i40evf_fill_default_direct_cmd_desc(&desc,
+						    i40e_aqc_opc_set_rss_lut);
+	else
+		i40evf_fill_default_direct_cmd_desc(&desc,
+						    i40e_aqc_opc_get_rss_lut);
+
+	/* Indirect command */
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+
+	cmd_resp->vsi_id =
+			cpu_to_le16((u16)((vsi_id <<
+					  I40E_AQC_SET_RSS_LUT_VSI_ID_SHIFT) &
+					  I40E_AQC_SET_RSS_LUT_VSI_ID_MASK));
+	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_LUT_VSI_VALID);
+
+	if (pf_lut)
+		cmd_resp->flags |= cpu_to_le16((u16)
+					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_PF <<
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+	else
+		cmd_resp->flags |= cpu_to_le16((u16)
+					((I40E_AQC_SET_RSS_LUT_TABLE_TYPE_VSI <<
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_SHIFT) &
+					I40E_AQC_SET_RSS_LUT_TABLE_TYPE_MASK));
+
+	cmd_resp->addr_high = cpu_to_le32(high_16_bits((u64)lut));
+	cmd_resp->addr_low = cpu_to_le32(lower_32_bits((u64)lut));
+
+	status = i40evf_asq_send_command(hw, &desc, lut, lut_size, NULL);
+
+	return status;
+}
+
+/**
+ * i40evf_aq_get_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ *
+ * get the RSS lookup table, PF or VSI type
+ **/
+i40e_status i40evf_aq_get_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+				  bool pf_lut, u8 *lut, u16 lut_size)
+{
+	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size,
+				       false);
+}
+
+/**
+ * i40evf_aq_set_rss_lut
+ * @hw: pointer to the hardware structure
+ * @vsi_id: vsi fw index
+ * @pf_lut: for PF table set true, for VSI table set false
+ * @lut: pointer to the lut buffer provided by the caller
+ * @lut_size: size of the lut buffer
+ *
+ * set the RSS lookup table, PF or VSI type
+ **/
+i40e_status i40evf_aq_set_rss_lut(struct i40e_hw *hw, u16 vsi_id,
+				  bool pf_lut, u8 *lut, u16 lut_size)
+{
+	return i40e_aq_get_set_rss_lut(hw, vsi_id, pf_lut, lut, lut_size, true);
+}
+
+/**
+ * i40e_aq_get_set_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ * @set: set true to set the key, false to get the key
+ *
+ * get the RSS key per VSI
+ **/
+static i40e_status i40e_aq_get_set_rss_key(struct i40e_hw *hw,
+				      u16 vsi_id,
+				      struct i40e_aqc_get_set_rss_key_data *key,
+				      bool set)
+{
+	i40e_status status;
+	struct i40e_aq_desc desc;
+	struct i40e_aqc_get_set_rss_key *cmd_resp =
+			(struct i40e_aqc_get_set_rss_key *)&desc.params.raw;
+	u16 key_size = sizeof(struct i40e_aqc_get_set_rss_key_data);
+
+	if (set)
+		i40evf_fill_default_direct_cmd_desc(&desc,
+						    i40e_aqc_opc_set_rss_key);
+	else
+		i40evf_fill_default_direct_cmd_desc(&desc,
+						    i40e_aqc_opc_get_rss_key);
+
+	/* Indirect command */
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_BUF);
+	desc.flags |= cpu_to_le16((u16)I40E_AQ_FLAG_RD);
+
+	cmd_resp->vsi_id =
+			cpu_to_le16((u16)((vsi_id <<
+					  I40E_AQC_SET_RSS_KEY_VSI_ID_SHIFT) &
+					  I40E_AQC_SET_RSS_KEY_VSI_ID_MASK));
+	cmd_resp->vsi_id |= cpu_to_le16((u16)I40E_AQC_SET_RSS_KEY_VSI_VALID);
+	cmd_resp->addr_high = cpu_to_le32(high_16_bits((u64)key));
+	cmd_resp->addr_low = cpu_to_le32(lower_32_bits((u64)key));
+
+	status = i40evf_asq_send_command(hw, &desc, key, key_size, NULL);
+
+	return status;
+}
+
+/**
+ * i40evf_aq_get_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ *
+ **/
+i40e_status i40evf_aq_get_rss_key(struct i40e_hw *hw,
+				  u16 vsi_id,
+				  struct i40e_aqc_get_set_rss_key_data *key)
+{
+	return i40e_aq_get_set_rss_key(hw, vsi_id, key, false);
+}
+
+/**
+ * i40evf_aq_set_rss_key
+ * @hw: pointer to the hw struct
+ * @vsi_id: vsi fw index
+ * @key: pointer to key info struct
+ *
+ * set the RSS key per VSI
+ **/
+i40e_status i40evf_aq_set_rss_key(struct i40e_hw *hw,
+				  u16 vsi_id,
+				  struct i40e_aqc_get_set_rss_key_data *key)
+{
+	return i40e_aq_get_set_rss_key(hw, vsi_id, key, true);
+}
+
 
 /* The i40evf_ptype_lookup table is used to convert from the 8-bit ptype in the
  * hardware to a bit-field that can be used by SW to more easily determine the

drivers/net/ethernet/intel/i40evf/i40e_prototype.h

@@ -63,6 +63,17 @@ i40e_status i40evf_aq_queue_shutdown(struct i40e_hw *hw, bool unloading);
 char *i40evf_aq_str(struct i40e_hw *hw, enum i40e_admin_queue_err aq_err);
 char *i40evf_stat_str(struct i40e_hw *hw, i40e_status stat_err);
 
+i40e_status i40evf_aq_get_rss_lut(struct i40e_hw *hw, u16 seid,
+				  bool pf_lut, u8 *lut, u16 lut_size);
+i40e_status i40evf_aq_set_rss_lut(struct i40e_hw *hw, u16 seid,
+				  bool pf_lut, u8 *lut, u16 lut_size);
+i40e_status i40evf_aq_get_rss_key(struct i40e_hw *hw,
+				  u16 seid,
+				  struct i40e_aqc_get_set_rss_key_data *key);
+i40e_status i40evf_aq_set_rss_key(struct i40e_hw *hw,
+				  u16 seid,
+				  struct i40e_aqc_get_set_rss_key_data *key);
+
 i40e_status i40e_set_mac_type(struct i40e_hw *hw);
 
 extern struct i40e_rx_ptype_decoded i40evf_ptype_lookup[];

drivers/net/ethernet/intel/i40evf/i40e_register.h

@@ -3366,4 +3366,64 @@
 #define I40E_VFQF_HREGION_OVERRIDE_ENA_7_MASK I40E_MASK(0x1, I40E_VFQF_HREGION_OVERRIDE_ENA_7_SHIFT)
 #define I40E_VFQF_HREGION_REGION_7_SHIFT 29
 #define I40E_VFQF_HREGION_REGION_7_MASK I40E_MASK(0x7, I40E_VFQF_HREGION_REGION_7_SHIFT)
-#endif
+#define I40E_VFINT_DYN_CTL01_WB_ON_ITR_SHIFT 30
+#define I40E_VFINT_DYN_CTL01_WB_ON_ITR_MASK I40E_MASK(0x1, I40E_VFINT_DYN_CTL01_WB_ON_ITR_SHIFT)
+#define I40E_VFINT_DYN_CTLN1_WB_ON_ITR_SHIFT 30
+#define I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK I40E_MASK(0x1, I40E_VFINT_DYN_CTLN1_WB_ON_ITR_SHIFT)
+#define I40E_VFPE_AEQALLOC1 0x0000A400 /* Reset: VFR */
+#define I40E_VFPE_AEQALLOC1_AECOUNT_SHIFT 0
+#define I40E_VFPE_AEQALLOC1_AECOUNT_MASK I40E_MASK(0xFFFFFFFF, I40E_VFPE_AEQALLOC1_AECOUNT_SHIFT)
+#define I40E_VFPE_CCQPHIGH1 0x00009800 /* Reset: VFR */
+#define I40E_VFPE_CCQPHIGH1_PECCQPHIGH_SHIFT 0
+#define I40E_VFPE_CCQPHIGH1_PECCQPHIGH_MASK I40E_MASK(0xFFFFFFFF, I40E_VFPE_CCQPHIGH1_PECCQPHIGH_SHIFT)
+#define I40E_VFPE_CCQPLOW1 0x0000AC00 /* Reset: VFR */
+#define I40E_VFPE_CCQPLOW1_PECCQPLOW_SHIFT 0
+#define I40E_VFPE_CCQPLOW1_PECCQPLOW_MASK I40E_MASK(0xFFFFFFFF, I40E_VFPE_CCQPLOW1_PECCQPLOW_SHIFT)
+#define I40E_VFPE_CCQPSTATUS1 0x0000B800 /* Reset: VFR */
+#define I40E_VFPE_CCQPSTATUS1_CCQP_DONE_SHIFT 0
+#define I40E_VFPE_CCQPSTATUS1_CCQP_DONE_MASK I40E_MASK(0x1, I40E_VFPE_CCQPSTATUS1_CCQP_DONE_SHIFT)
+#define I40E_VFPE_CCQPSTATUS1_HMC_PROFILE_SHIFT 4
+#define I40E_VFPE_CCQPSTATUS1_HMC_PROFILE_MASK I40E_MASK(0x7, I40E_VFPE_CCQPSTATUS1_HMC_PROFILE_SHIFT)
+#define I40E_VFPE_CCQPSTATUS1_RDMA_EN_VFS_SHIFT 16
+#define I40E_VFPE_CCQPSTATUS1_RDMA_EN_VFS_MASK I40E_MASK(0x3F, I40E_VFPE_CCQPSTATUS1_RDMA_EN_VFS_SHIFT)
+#define I40E_VFPE_CCQPSTATUS1_CCQP_ERR_SHIFT 31
+#define I40E_VFPE_CCQPSTATUS1_CCQP_ERR_MASK I40E_MASK(0x1, I40E_VFPE_CCQPSTATUS1_CCQP_ERR_SHIFT)
+#define I40E_VFPE_CQACK1 0x0000B000 /* Reset: VFR */
+#define I40E_VFPE_CQACK1_PECQID_SHIFT 0
+#define I40E_VFPE_CQACK1_PECQID_MASK I40E_MASK(0x1FFFF, I40E_VFPE_CQACK1_PECQID_SHIFT)
+#define I40E_VFPE_CQARM1 0x0000B400 /* Reset: VFR */
+#define I40E_VFPE_CQARM1_PECQID_SHIFT 0
+#define I40E_VFPE_CQARM1_PECQID_MASK I40E_MASK(0x1FFFF, I40E_VFPE_CQARM1_PECQID_SHIFT)
+#define I40E_VFPE_CQPDB1 0x0000BC00 /* Reset: VFR */
+#define I40E_VFPE_CQPDB1_WQHEAD_SHIFT 0
+#define I40E_VFPE_CQPDB1_WQHEAD_MASK I40E_MASK(0x7FF, I40E_VFPE_CQPDB1_WQHEAD_SHIFT)
+#define I40E_VFPE_CQPERRCODES1 0x00009C00 /* Reset: VFR */
+#define I40E_VFPE_CQPERRCODES1_CQP_MINOR_CODE_SHIFT 0
+#define I40E_VFPE_CQPERRCODES1_CQP_MINOR_CODE_MASK I40E_MASK(0xFFFF, I40E_VFPE_CQPERRCODES1_CQP_MINOR_CODE_SHIFT)
+#define I40E_VFPE_CQPERRCODES1_CQP_MAJOR_CODE_SHIFT 16
+#define I40E_VFPE_CQPERRCODES1_CQP_MAJOR_CODE_MASK I40E_MASK(0xFFFF, I40E_VFPE_CQPERRCODES1_CQP_MAJOR_CODE_SHIFT)
+#define I40E_VFPE_CQPTAIL1 0x0000A000 /* Reset: VFR */
+#define I40E_VFPE_CQPTAIL1_WQTAIL_SHIFT 0
+#define I40E_VFPE_CQPTAIL1_WQTAIL_MASK I40E_MASK(0x7FF, I40E_VFPE_CQPTAIL1_WQTAIL_SHIFT)
+#define I40E_VFPE_CQPTAIL1_CQP_OP_ERR_SHIFT 31
+#define I40E_VFPE_CQPTAIL1_CQP_OP_ERR_MASK I40E_MASK(0x1, I40E_VFPE_CQPTAIL1_CQP_OP_ERR_SHIFT)
+#define I40E_VFPE_IPCONFIG01 0x00008C00 /* Reset: VFR */
+#define I40E_VFPE_IPCONFIG01_PEIPID_SHIFT 0
+#define I40E_VFPE_IPCONFIG01_PEIPID_MASK I40E_MASK(0xFFFF, I40E_VFPE_IPCONFIG01_PEIPID_SHIFT)
+#define I40E_VFPE_IPCONFIG01_USEENTIREIDRANGE_SHIFT 16
+#define I40E_VFPE_IPCONFIG01_USEENTIREIDRANGE_MASK I40E_MASK(0x1, I40E_VFPE_IPCONFIG01_USEENTIREIDRANGE_SHIFT)
+#define I40E_VFPE_MRTEIDXMASK1 0x00009000 /* Reset: VFR */
+#define I40E_VFPE_MRTEIDXMASK1_MRTEIDXMASKBITS_SHIFT 0
+#define I40E_VFPE_MRTEIDXMASK1_MRTEIDXMASKBITS_MASK I40E_MASK(0x1F, I40E_VFPE_MRTEIDXMASK1_MRTEIDXMASKBITS_SHIFT)
+#define I40E_VFPE_RCVUNEXPECTEDERROR1 0x00009400 /* Reset: VFR */
+#define I40E_VFPE_RCVUNEXPECTEDERROR1_TCP_RX_UNEXP_ERR_SHIFT 0
+#define I40E_VFPE_RCVUNEXPECTEDERROR1_TCP_RX_UNEXP_ERR_MASK I40E_MASK(0xFFFFFF, I40E_VFPE_RCVUNEXPECTEDERROR1_TCP_RX_UNEXP_ERR_SHIFT)
+#define I40E_VFPE_TCPNOWTIMER1 0x0000A800 /* Reset: VFR */
+#define I40E_VFPE_TCPNOWTIMER1_TCP_NOW_SHIFT 0
+#define I40E_VFPE_TCPNOWTIMER1_TCP_NOW_MASK I40E_MASK(0xFFFFFFFF, I40E_VFPE_TCPNOWTIMER1_TCP_NOW_SHIFT)
+#define I40E_VFPE_WQEALLOC1 0x0000C000 /* Reset: VFR */
+#define I40E_VFPE_WQEALLOC1_PEQPID_SHIFT 0
+#define I40E_VFPE_WQEALLOC1_PEQPID_MASK I40E_MASK(0x3FFFF, I40E_VFPE_WQEALLOC1_PEQPID_SHIFT)
+#define I40E_VFPE_WQEALLOC1_WQE_DESC_INDEX_SHIFT 20
+#define I40E_VFPE_WQEALLOC1_WQE_DESC_INDEX_MASK I40E_MASK(0xFFF, I40E_VFPE_WQEALLOC1_WQE_DESC_INDEX_SHIFT)
+#endif /* _I40E_REGISTER_H_ */

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

@@ -366,15 +366,32 @@ static bool i40e_clean_tx_irq(struct i40e_ring *tx_ring, int budget)
  **/
 static void i40e_force_wb(struct i40e_vsi *vsi, struct i40e_q_vector *q_vector)
 {
-	u32 val = I40E_VFINT_DYN_CTLN_INTENA_MASK |
+	u16 flags = q_vector->tx.ring[0].flags;
+
+	if (flags & I40E_TXR_FLAGS_WB_ON_ITR) {
+		u32 val;
+
+		if (q_vector->arm_wb_state)
+			return;
+
+		val = I40E_VFINT_DYN_CTLN1_WB_ON_ITR_MASK;
+
+		wr32(&vsi->back->hw,
+		     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
+					  vsi->base_vector - 1),
+		     val);
+		q_vector->arm_wb_state = true;
+	} else {
+		u32 val = I40E_VFINT_DYN_CTLN1_INTENA_MASK |
 			  I40E_VFINT_DYN_CTLN1_ITR_INDX_MASK | /* set noitr */
-		  I40E_VFINT_DYN_CTLN_SWINT_TRIG_MASK |
-		  I40E_VFINT_DYN_CTLN_SW_ITR_INDX_ENA_MASK;
+			  I40E_VFINT_DYN_CTLN1_SWINT_TRIG_MASK |
+			  I40E_VFINT_DYN_CTLN1_SW_ITR_INDX_ENA_MASK;
 			  /* allow 00 to be written to the index */
 
 		wr32(&vsi->back->hw,
-	     I40E_VFINT_DYN_CTLN1(q_vector->v_idx + vsi->base_vector - 1),
-	     val);
+		     I40E_VFINT_DYN_CTLN1(q_vector->v_idx +
+					  vsi->base_vector - 1), val);
+	}
 }
 
 /**
@@ -1372,6 +1389,9 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget)
 		return budget;
 	}
 
+	if (vsi->back->flags & I40E_TXR_FLAGS_WB_ON_ITR)
+		q_vector->arm_wb_state = false;
+
 	/* Work is done so exit the polling mode and re-enable the interrupt */
 	napi_complete(napi);
 	i40e_update_enable_itr(vsi, q_vector);
@@ -1508,11 +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;
@@ -1551,6 +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

@@ -78,6 +78,18 @@ enum i40e_dyn_idx_t {
 	BIT_ULL(I40E_FILTER_PCTYPE_FRAG_IPV6) | \
 	BIT_ULL(I40E_FILTER_PCTYPE_L2_PAYLOAD))
 
+#define I40E_DEFAULT_RSS_HENA_EXPANDED (I40E_DEFAULT_RSS_HENA | \
+		BIT(I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK) | \
+		BIT(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP) | \
+		BIT(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP) | \
+		BIT(I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK) | \
+		BIT(I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP) | \
+		BIT(I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP))
+
+#define i40e_pf_get_default_rss_hena(pf) \
+	(((pf)->flags & I40E_FLAG_MULTIPLE_TCP_UDP_RSS_PCTYPE) ? \
+		I40E_DEFAULT_RSS_HENA_EXPANDED : I40E_DEFAULT_RSS_HENA)
+
 /* Supported Rx Buffer Sizes */
 #define I40E_RXBUFFER_512   512    /* Used for packet split */
 #define I40E_RXBUFFER_2048  2048
@@ -250,6 +262,10 @@ struct i40e_ring {
 	bool ring_active;		/* is ring online or not */
 	bool arm_wb;		/* do something to arm write back */
 
+	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

@@ -47,6 +47,11 @@
 #define I40E_DEV_ID_20G_KR2		0x1587
 #define I40E_DEV_ID_VF			0x154C
 #define I40E_DEV_ID_VF_HV		0x1571
+#define I40E_DEV_ID_SFP_X722		0x37D0
+#define I40E_DEV_ID_1G_BASE_T_X722	0x37D1
+#define I40E_DEV_ID_10G_BASE_T_X722	0x37D2
+#define I40E_DEV_ID_X722_VF		0x37CD
+#define I40E_DEV_ID_X722_VF_HV		0x37D9
 
 #define i40e_is_40G_device(d)		((d) == I40E_DEV_ID_QSFP_A  || \
 					 (d) == I40E_DEV_ID_QSFP_B  || \
@@ -120,6 +125,8 @@ enum i40e_mac_type {
 	I40E_MAC_X710,
 	I40E_MAC_XL710,
 	I40E_MAC_VF,
+	I40E_MAC_X722,
+	I40E_MAC_X722_VF,
 	I40E_MAC_GENERIC,
 };
 
@@ -496,7 +503,8 @@ struct i40e_hw {
 
 static inline bool i40e_is_vf(struct i40e_hw *hw)
 {
-	return hw->mac.type == I40E_MAC_VF;
+	return (hw->mac.type == I40E_MAC_VF ||
+		hw->mac.type == I40E_MAC_X722_VF);
 }
 
 struct i40e_driver_version {
@@ -593,14 +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!!! */
 };
 
@@ -941,6 +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;
@@ -959,15 +973,24 @@ struct i40e_filter_program_desc {
 
 /* Packet Classifier Types for filters */
 enum i40e_filter_pctype {
-	/* Note: Values 0-30 are reserved for future use */
+	/* Note: Values 0-28 are reserved for future use.
+	 * Value 29, 30, 32 are not supported on XL710 and X710.
+	 */
+	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV4_UDP	= 29,
+	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV4_UDP	= 30,
 	I40E_FILTER_PCTYPE_NONF_IPV4_UDP		= 31,
-	/* Note: Value 32 is reserved for future use */
+	I40E_FILTER_PCTYPE_NONF_IPV4_TCP_SYN_NO_ACK	= 32,
 	I40E_FILTER_PCTYPE_NONF_IPV4_TCP		= 33,
 	I40E_FILTER_PCTYPE_NONF_IPV4_SCTP		= 34,
 	I40E_FILTER_PCTYPE_NONF_IPV4_OTHER		= 35,
 	I40E_FILTER_PCTYPE_FRAG_IPV4			= 36,
-	/* Note: Values 37-40 are reserved for future use */
+	/* Note: Values 37-38 are reserved for future use.
+	 * Value 39, 40, 42 are not supported on XL710 and X710.
+	 */
+	I40E_FILTER_PCTYPE_NONF_UNICAST_IPV6_UDP	= 39,
+	I40E_FILTER_PCTYPE_NONF_MULTICAST_IPV6_UDP	= 40,
 	I40E_FILTER_PCTYPE_NONF_IPV6_UDP		= 41,
+	I40E_FILTER_PCTYPE_NONF_IPV6_TCP_SYN_NO_ACK	= 42,
 	I40E_FILTER_PCTYPE_NONF_IPV6_TCP		= 43,
 	I40E_FILTER_PCTYPE_NONF_IPV6_SCTP		= 44,
 	I40E_FILTER_PCTYPE_NONF_IPV6_OTHER		= 45,

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

@@ -101,6 +101,8 @@ struct i40e_vsi {
 #define MAX_RX_QUEUES 8
 #define MAX_TX_QUEUES MAX_RX_QUEUES
 
+#define I40EVF_HKEY_ARRAY_SIZE ((I40E_VFQF_HKEY_MAX_INDEX + 1) * 4)
+
 /* MAX_MSIX_Q_VECTORS of these are allocated,
  * but we only use one per queue-specific vector.
  */
@@ -115,6 +117,7 @@ struct i40e_q_vector {
 	u8 num_ringpairs;	/* total number of ring pairs in vector */
 	int v_idx;	  /* vector index in list */
 	char name[IFNAMSIZ + 9];
+	bool arm_wb_state;
 	cpumask_var_t affinity_mask;
 };
 
@@ -218,11 +221,15 @@ struct i40evf_adapter {
 #define I40EVF_FLAG_PF_COMMS_FAILED              BIT(8)
 #define I40EVF_FLAG_RESET_PENDING                BIT(9)
 #define I40EVF_FLAG_RESET_NEEDED                 BIT(10)
-/* duplcates for common code */
+#define I40EVF_FLAG_WB_ON_ITR_CAPABLE		BIT(11)
+#define I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE	BIT(12)
+/* duplicates for common code */
 #define I40E_FLAG_FDIR_ATR_ENABLED		 0
 #define I40E_FLAG_DCB_ENABLED			 0
 #define I40E_FLAG_IN_NETPOLL			 I40EVF_FLAG_IN_NETPOLL
 #define I40E_FLAG_RX_CSUM_ENABLED                I40EVF_FLAG_RX_CSUM_ENABLED
+#define I40E_FLAG_WB_ON_ITR_CAPABLE		I40EVF_FLAG_WB_ON_ITR_CAPABLE
+#define I40E_FLAG_OUTER_UDP_CSUM_CAPABLE	I40EVF_FLAG_OUTER_UDP_CSUM_CAPABLE
 	/* flags for admin queue service task */
 	u32 aq_required;
 #define I40EVF_FLAG_AQ_ENABLE_QUEUES		BIT(0)
@@ -234,6 +241,7 @@ 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_GET_CONFIG		BIT(10)
 
 	/* OS defined structs */

drivers/net/ethernet/intel/i40evf/i40evf_main.c

@@ -49,6 +49,7 @@ static const char i40evf_copyright[] =
  */
 static const struct pci_device_id i40evf_pci_tbl[] = {
 	{PCI_VDEVICE(INTEL, I40E_DEV_ID_VF), 0},
+	{PCI_VDEVICE(INTEL, I40E_DEV_ID_X722_VF), 0},
 	/* required last entry */
 	{0, }
 };
@@ -1171,6 +1172,113 @@ static int i40evf_set_interrupt_capability(struct i40evf_adapter *adapter)
 	return err;
 }
 
+/**
+ * i40e_configure_rss_aq - Prepare for RSS using AQ commands
+ * @vsi: vsi structure
+ * @seed: RSS hash seed
+ **/
+static void i40evf_configure_rss_aq(struct i40e_vsi *vsi, const u8 *seed)
+{
+	struct i40e_aqc_get_set_rss_key_data rss_key;
+	struct i40evf_adapter *adapter = vsi->back;
+	struct i40e_hw *hw = &adapter->hw;
+	int ret = 0, i;
+	u8 *rss_lut;
+
+	if (!vsi->id)
+		return;
+
+	if (adapter->current_op != I40E_VIRTCHNL_OP_UNKNOWN) {
+		/* bail because we already have a command pending */
+		dev_err(&adapter->pdev->dev, "Cannot confiure RSS, command %d pending\n",
+			adapter->current_op);
+		return;
+	}
+
+	memset(&rss_key, 0, sizeof(rss_key));
+	memcpy(&rss_key, seed, sizeof(rss_key));
+
+	rss_lut = kzalloc(((I40E_VFQF_HLUT_MAX_INDEX + 1) * 4), GFP_KERNEL);
+	if (!rss_lut)
+		return;
+
+	/* Populate the LUT with max no. PF queues in round robin fashion */
+	for (i = 0; i <= (I40E_VFQF_HLUT_MAX_INDEX * 4); i++)
+		rss_lut[i] = i % adapter->num_active_queues;
+
+	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 = i40evf_aq_set_rss_lut(hw, vsi->id, false, rss_lut,
+				    (I40E_VFQF_HLUT_MAX_INDEX + 1) * 4);
+	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));
+}
+
+/**
+ * i40e_configure_rss_reg - Prepare for RSS if used
+ * @adapter: board private structure
+ * @seed: RSS hash seed
+ **/
+static void i40evf_configure_rss_reg(struct i40evf_adapter *adapter,
+				     const u8 *seed)
+{
+	struct i40e_hw *hw = &adapter->hw;
+	u32 *seed_dw = (u32 *)seed;
+	u32 cqueue = 0;
+	u32 lut = 0;
+	int i, j;
+
+	/* Fill out hash function seed */
+	for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++)
+		wr32(hw, I40E_VFQF_HKEY(i), seed_dw[i]);
+
+	/* Populate the LUT with max no. PF queues in round robin fashion */
+	for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) {
+		lut = 0;
+		for (j = 0; j < 4; j++) {
+			if (cqueue == adapter->num_active_queues)
+				cqueue = 0;
+			lut |= ((cqueue) << (8 * j));
+			cqueue++;
+		}
+		wr32(hw, I40E_VFQF_HLUT(i), lut);
+	}
+	i40e_flush(hw);
+}
+
+/**
+ * i40evf_configure_rss - Prepare for RSS
+ * @adapter: board private structure
+ **/
+static void i40evf_configure_rss(struct i40evf_adapter *adapter)
+{
+	struct i40e_hw *hw = &adapter->hw;
+	u8 seed[I40EVF_HKEY_ARRAY_SIZE];
+	u64 hena;
+
+	netdev_rss_key_fill((void *)seed, I40EVF_HKEY_ARRAY_SIZE);
+
+	/* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */
+	hena = I40E_DEFAULT_RSS_HENA;
+	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
+	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
+
+	if (RSS_AQ(adapter))
+		i40evf_configure_rss_aq(&adapter->vsi, seed);
+	else
+		i40evf_configure_rss_reg(adapter, seed);
+}
+
 /**
  * i40evf_alloc_q_vectors - Allocate memory for interrupt vectors
  * @adapter: board private structure to initialize
@@ -1416,6 +1524,16 @@ static void i40evf_watchdog_task(struct work_struct *work)
 		goto watchdog_done;
 	}
 
+	if (adapter->aq_required & I40EVF_FLAG_AQ_CONFIGURE_RSS) {
+		/* This message goes straight to the firmware, not the
+		 * PF, so we don't have to set current_op as we will
+		 * not get a response through the ARQ.
+		 */
+		i40evf_configure_rss(adapter);
+		adapter->aq_required &= ~I40EVF_FLAG_AQ_CONFIGURE_RSS;
+		goto watchdog_done;
+	}
+
 	if (adapter->state == __I40EVF_RUNNING)
 		i40evf_request_stats(adapter);
 watchdog_done:
@@ -1438,45 +1556,6 @@ static void i40evf_watchdog_task(struct work_struct *work)
 	schedule_work(&adapter->adminq_task);
 }
 
-/**
- * i40evf_configure_rss - Prepare for RSS
- * @adapter: board private structure
- **/
-static void i40evf_configure_rss(struct i40evf_adapter *adapter)
-{
-	u32 rss_key[I40E_VFQF_HKEY_MAX_INDEX + 1];
-	struct i40e_hw *hw = &adapter->hw;
-	u32 cqueue = 0;
-	u32 lut = 0;
-	int i, j;
-	u64 hena;
-
-	/* Hash type is configured by the PF - we just supply the key */
-	netdev_rss_key_fill(rss_key, sizeof(rss_key));
-
-	/* Fill out hash function seed */
-	for (i = 0; i <= I40E_VFQF_HKEY_MAX_INDEX; i++)
-		wr32(hw, I40E_VFQF_HKEY(i), rss_key[i]);
-
-	/* Enable PCTYPES for RSS, TCP/UDP with IPv4/IPv6 */
-	hena = I40E_DEFAULT_RSS_HENA;
-	wr32(hw, I40E_VFQF_HENA(0), (u32)hena);
-	wr32(hw, I40E_VFQF_HENA(1), (u32)(hena >> 32));
-
-	/* Populate the LUT with max no. of queues in round robin fashion */
-	for (i = 0; i <= I40E_VFQF_HLUT_MAX_INDEX; i++) {
-		lut = 0;
-		for (j = 0; j < 4; j++) {
-			if (cqueue == adapter->num_active_queues)
-				cqueue = 0;
-			lut |= ((cqueue) << (8 * j));
-			cqueue++;
-		}
-		wr32(hw, I40E_VFQF_HLUT(i), lut);
-	}
-	i40e_flush(hw);
-}
-
 #define I40EVF_RESET_WAIT_MS 10
 #define I40EVF_RESET_WAIT_COUNT 500
 /**
@@ -2186,6 +2265,7 @@ static void i40evf_init_task(struct work_struct *work)
 	if (err)
 		goto err_sw_init;
 	i40evf_map_rings_to_vectors(adapter);
+	if (!RSS_AQ(adapter))
 		i40evf_configure_rss(adapter);
 	err = i40evf_request_misc_irq(adapter);
 	if (err)
@@ -2211,6 +2291,13 @@ static void i40evf_init_task(struct work_struct *work)
 	adapter->state = __I40EVF_DOWN;
 	set_bit(__I40E_DOWN, &adapter->vsi.state);
 	i40evf_misc_irq_enable(adapter);
+
+	if (RSS_AQ(adapter)) {
+		adapter->aq_required |= I40EVF_FLAG_AQ_CONFIGURE_RSS;
+		mod_timer_pending(&adapter->watchdog_timer, jiffies + 1);
+	} else {
+		i40evf_configure_rss(adapter);
+	}
 	return;
 restart:
 	schedule_delayed_work(&adapter->init_task,