1 files changed, 119 insertions, 136 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 321d8be80871..f95ce9b5e4fb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -47,8 +47,8 @@ static const char i40e_driver_string[] =
 #define DRV_KERN "-k"
 
 #define DRV_VERSION_MAJOR 2
-#define DRV_VERSION_MINOR 1
-#define DRV_VERSION_BUILD 14
+#define DRV_VERSION_MINOR 3
+#define DRV_VERSION_BUILD 2
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
@@ -1573,11 +1573,18 @@ static int i40e_set_mac(struct net_device *netdev, void *p)
 	else
 		netdev_info(netdev, "set new mac address %pM\n", addr->sa_data);
 
+	/* Copy the address first, so that we avoid a possible race with
+	 * .set_rx_mode(). If we copy after changing the address in the filter
+	 * list, we might open ourselves to a narrow race window where
+	 * .set_rx_mode could delete our dev_addr filter and prevent traffic
+	 * from passing.
+	 */
+	ether_addr_copy(netdev->dev_addr, addr->sa_data);
+
 	spin_lock_bh(&vsi->mac_filter_hash_lock);
 	i40e_del_mac_filter(vsi, netdev->dev_addr);
 	i40e_add_mac_filter(vsi, addr->sa_data);
 	spin_unlock_bh(&vsi->mac_filter_hash_lock);
-	ether_addr_copy(netdev->dev_addr, addr->sa_data);
 	if (vsi->type == I40E_VSI_MAIN) {
 		i40e_status ret;
 
@@ -1811,6 +1818,10 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
 	num_tc_qps = qcount / numtc;
 	num_tc_qps = min_t(int, num_tc_qps, i40e_pf_get_max_q_per_tc(pf));
 
+	/* Do not allow use more TC queue pairs than MSI-X vectors exist */
+	if (pf->flags & I40E_FLAG_MSIX_ENABLED)
+		num_tc_qps = min_t(int, num_tc_qps, pf->num_lan_msix);
+
 	/* Setup queue offset/count for all TCs for given VSI */
 	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
 		/* See if the given TC is enabled for the given VSI */
@@ -1923,6 +1934,14 @@ static int i40e_addr_unsync(struct net_device *netdev, const u8 *addr)
 	struct i40e_netdev_priv *np = netdev_priv(netdev);
 	struct i40e_vsi *vsi = np->vsi;
 
+	/* Under some circumstances, we might receive a request to delete
+	 * our own device address from our uc list. Because we store the
+	 * device address in the VSI's MAC/VLAN filter list, we need to ignore
+	 * such requests and not delete our device address from this list.
+	 */
+	if (ether_addr_equal(addr, netdev->dev_addr))
+		return 0;
+
 	i40e_del_mac_filter(vsi, addr);
 
 	return 0;
@@ -4107,6 +4126,7 @@ static void i40e_vsi_map_rings_to_vectors(struct i40e_vsi *vsi)
 		num_ringpairs = DIV_ROUND_UP(qp_remaining, q_vectors - v_start);
 
 		q_vector->num_ringpairs = num_ringpairs;
+		q_vector->reg_idx = q_vector->v_idx + vsi->base_vector - 1;
 
 		q_vector->rx.count = 0;
 		q_vector->tx.count = 0;
@@ -4862,104 +4882,6 @@ static int i40e_pf_wait_queues_disabled(struct i40e_pf *pf)
 #endif
 
 /**
- * i40e_detect_recover_hung_queue - Function to detect and recover hung_queue
- * @q_idx: TX queue number
- * @vsi: Pointer to VSI struct
- *
- * This function checks specified queue for given VSI. Detects hung condition.
- * We proactively detect hung TX queues by checking if interrupts are disabled
- * but there are pending descriptors.  If it appears hung, attempt to recover
- * by triggering a SW interrupt.
- **/
-static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
-{
-	struct i40e_ring *tx_ring = NULL;
-	struct i40e_pf	*pf;
-	u32 val, tx_pending;
-	int i;
-
-	pf = vsi->back;
-
-	/* now that we have an index, find the tx_ring struct */
-	for (i = 0; i < vsi->num_queue_pairs; i++) {
-		if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
-			if (q_idx == vsi->tx_rings[i]->queue_index) {
-				tx_ring = vsi->tx_rings[i];
-				break;
-			}
-		}
-	}
-
-	if (!tx_ring)
-		return;
-
-	/* Read interrupt register */
-	if (pf->flags & I40E_FLAG_MSIX_ENABLED)
-		val = rd32(&pf->hw,
-			   I40E_PFINT_DYN_CTLN(tx_ring->q_vector->v_idx +
-					       tx_ring->vsi->base_vector - 1));
-	else
-		val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);
-
-	tx_pending = i40e_get_tx_pending(tx_ring);
-
-	/* Interrupts are disabled and TX pending is non-zero,
-	 * trigger the SW interrupt (don't wait). Worst case
-	 * there will be one extra interrupt which may result
-	 * into not cleaning any queues because queues are cleaned.
-	 */
-	if (tx_pending && (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK)))
-		i40e_force_wb(vsi, tx_ring->q_vector);
-}
-
-/**
- * i40e_detect_recover_hung - Function to detect and recover hung_queues
- * @pf:  pointer to PF struct
- *
- * LAN VSI has netdev and netdev has TX queues. This function is to check
- * each of those TX queues if they are hung, trigger recovery by issuing
- * SW interrupt.
- **/
-static void i40e_detect_recover_hung(struct i40e_pf *pf)
-{
-	struct net_device *netdev;
-	struct i40e_vsi *vsi;
-	unsigned int i;
-
-	/* Only for LAN VSI */
-	vsi = pf->vsi[pf->lan_vsi];
-
-	if (!vsi)
-		return;
-
-	/* Make sure, VSI state is not DOWN/RECOVERY_PENDING */
-	if (test_bit(__I40E_VSI_DOWN, vsi->back->state) ||
-	    test_bit(__I40E_RESET_RECOVERY_PENDING, vsi->back->state))
-		return;
-
-	/* Make sure type is MAIN VSI */
-	if (vsi->type != I40E_VSI_MAIN)
-		return;
-
-	netdev = vsi->netdev;
-	if (!netdev)
-		return;
-
-	/* Bail out if netif_carrier is not OK */
-	if (!netif_carrier_ok(netdev))
-		return;
-
-	/* Go thru' TX queues for netdev */
-	for (i = 0; i < netdev->num_tx_queues; i++) {
-		struct netdev_queue *q;
-
-		q = netdev_get_tx_queue(netdev, i);
-		if (q)
-			i40e_detect_recover_hung_queue(i, vsi);
-	}
-}
-
-/**
  * i40e_get_iscsi_tc_map - Return TC map for iSCSI APP
  * @pf: pointer to PF
  *
@@ -5327,6 +5249,8 @@ static void i40e_vsi_update_queue_map(struct i40e_vsi *vsi,
 static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
 {
 	u8 bw_share[I40E_MAX_TRAFFIC_CLASS] = {0};
+	struct i40e_pf *pf = vsi->back;
+	struct i40e_hw *hw = &pf->hw;
 	struct i40e_vsi_context ctxt;
 	int ret = 0;
 	int i;
@@ -5344,10 +5268,40 @@ static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
 
 	ret = i40e_vsi_configure_bw_alloc(vsi, enabled_tc, bw_share);
 	if (ret) {
-		dev_info(&vsi->back->pdev->dev,
+		struct i40e_aqc_query_vsi_bw_config_resp bw_config = {0};
+
+		dev_info(&pf->pdev->dev,
 			 "Failed configuring TC map %d for VSI %d\n",
 			 enabled_tc, vsi->seid);
-		goto out;
+		ret = i40e_aq_query_vsi_bw_config(hw, vsi->seid,
+						  &bw_config, NULL);
+		if (ret) {
+			dev_info(&pf->pdev->dev,
+				 "Failed querying vsi bw info, err %s aq_err %s\n",
+				 i40e_stat_str(hw, ret),
+				 i40e_aq_str(hw, hw->aq.asq_last_status));
+			goto out;
+		}
+		if ((bw_config.tc_valid_bits & enabled_tc) != enabled_tc) {
+			u8 valid_tc = bw_config.tc_valid_bits & enabled_tc;
+
+			if (!valid_tc)
+				valid_tc = bw_config.tc_valid_bits;
+			/* Always enable TC0, no matter what */
+			valid_tc |= 1;
+			dev_info(&pf->pdev->dev,
+				 "Requested tc 0x%x, but FW reports 0x%x as valid. Attempting to use 0x%x.\n",
+				 enabled_tc, bw_config.tc_valid_bits, valid_tc);
+			enabled_tc = valid_tc;
+		}
+
+		ret = i40e_vsi_configure_bw_alloc(vsi, enabled_tc, bw_share);
+		if (ret) {
+			dev_err(&pf->pdev->dev,
+				"Unable to  configure TC map %d for VSI %d\n",
+				enabled_tc, vsi->seid);
+			goto out;
+		}
 	}
 
 	/* Update Queue Pairs Mapping for currently enabled UPs */
@@ -5387,13 +5341,12 @@ static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
 	/* Update the VSI after updating the VSI queue-mapping
 	 * information
 	 */
-	ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
+	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
 	if (ret) {
-		dev_info(&vsi->back->pdev->dev,
+		dev_info(&pf->pdev->dev,
 			 "Update vsi tc config failed, err %s aq_err %s\n",
-			 i40e_stat_str(&vsi->back->hw, ret),
-			 i40e_aq_str(&vsi->back->hw,
-				     vsi->back->hw.aq.asq_last_status));
+			 i40e_stat_str(hw, ret),
+			 i40e_aq_str(hw, hw->aq.asq_last_status));
 		goto out;
 	}
 	/* update the local VSI info with updated queue map */
@@ -5403,11 +5356,10 @@ static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
 	/* Update current VSI BW information */
 	ret = i40e_vsi_get_bw_info(vsi);
 	if (ret) {
-		dev_info(&vsi->back->pdev->dev,
+		dev_info(&pf->pdev->dev,
 			 "Failed updating vsi bw info, err %s aq_err %s\n",
-			 i40e_stat_str(&vsi->back->hw, ret),
-			 i40e_aq_str(&vsi->back->hw,
-				     vsi->back->hw.aq.asq_last_status));
+			 i40e_stat_str(hw, ret),
+			 i40e_aq_str(hw, hw->aq.asq_last_status));
 		goto out;
 	}
 
@@ -6038,8 +5990,8 @@ static int i40e_validate_and_set_switch_mode(struct i40e_vsi *vsi)
 	/* Set Bit 7 to be valid */
 	mode = I40E_AQ_SET_SWITCH_BIT7_VALID;
 
-	/* Set L4type to both TCP and UDP support */
-	mode |= I40E_AQ_SET_SWITCH_L4_TYPE_BOTH;
+	/* Set L4type for TCP support */
+	mode |= I40E_AQ_SET_SWITCH_L4_TYPE_TCP;
 
 	/* Set cloud filter mode */
 	mode |= I40E_AQ_SET_SWITCH_MODE_NON_TUNNEL;
@@ -6373,8 +6325,11 @@ static int i40e_init_pf_dcb(struct i40e_pf *pf)
 	struct i40e_hw *hw = &pf->hw;
 	int err = 0;
 
-	/* Do not enable DCB for SW1 and SW2 images even if the FW is capable */
-	if (pf->hw_features & I40E_HW_NO_DCB_SUPPORT)
+	/* Do not enable DCB for SW1 and SW2 images even if the FW is capable
+	 * Also do not enable DCBx if FW LLDP agent is disabled
+	 */
+	if ((pf->hw_features & I40E_HW_NO_DCB_SUPPORT) ||
+	    (pf->flags & I40E_FLAG_DISABLE_FW_LLDP))
 		goto out;
 
 	/* Get the initial DCB configuration */
@@ -6401,6 +6356,9 @@ static int i40e_init_pf_dcb(struct i40e_pf *pf)
 			dev_dbg(&pf->pdev->dev,
 				"DCBX offload is supported for this PF.\n");
 		}
+	} else if (pf->hw.aq.asq_last_status == I40E_AQ_RC_EPERM) {
+		dev_info(&pf->pdev->dev, "FW LLDP disabled for this PF.\n");
+		pf->flags |= I40E_FLAG_DISABLE_FW_LLDP;
 	} else {
 		dev_info(&pf->pdev->dev,
 			 "Query for DCB configuration failed, err %s aq_err %s\n",
@@ -6969,18 +6927,18 @@ static int i40e_add_del_cloud_filter_big_buf(struct i40e_vsi *vsi,
 	     is_valid_ether_addr(filter->src_mac)) ||
 	    (is_multicast_ether_addr(filter->dst_mac) &&
 	     is_multicast_ether_addr(filter->src_mac)))
-		return -EINVAL;
+		return -EOPNOTSUPP;
 
-	/* Make sure port is specified, otherwise bail out, for channel
-	 * specific cloud filter needs 'L4 port' to be non-zero
+	/* Big buffer cloud filter needs 'L4 port' to be non-zero. Also, UDP
+	 * ports are not supported via big buffer now.
 	 */
-	if (!filter->dst_port)
-		return -EINVAL;
+	if (!filter->dst_port || filter->ip_proto == IPPROTO_UDP)
+		return -EOPNOTSUPP;
 
 	/* adding filter using src_port/src_ip is not supported at this stage */
 	if (filter->src_port || filter->src_ipv4 ||
 	    !ipv6_addr_any(&filter->ip.v6.src_ip6))
-		return -EINVAL;
+		return -EOPNOTSUPP;
 
 	/* copy element needed to add cloud filter from filter */
 	i40e_set_cld_element(filter, &cld_filter.element);
@@ -6991,7 +6949,7 @@ static int i40e_add_del_cloud_filter_big_buf(struct i40e_vsi *vsi,
 	    is_multicast_ether_addr(filter->src_mac)) {
 		/* MAC + IP : unsupported mode */
 		if (filter->dst_ipv4)
-			return -EINVAL;
+			return -EOPNOTSUPP;
 
 		/* since we validated that L4 port must be valid before
 		 * we get here, start with respective "flags" value
@@ -7356,7 +7314,7 @@ static int i40e_configure_clsflower(struct i40e_vsi *vsi,
 
 	if (tc < 0) {
 		dev_err(&vsi->back->pdev->dev, "Invalid traffic class\n");
-		return -EINVAL;
+		return -EOPNOTSUPP;
 	}
 
 	if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state) ||
@@ -7490,9 +7448,6 @@ static int i40e_setup_tc_cls_flower(struct i40e_netdev_priv *np,
 {
 	struct i40e_vsi *vsi = np->vsi;
 
-	if (cls_flower->common.chain_index)
-		return -EOPNOTSUPP;
-
 	switch (cls_flower->command) {
 	case TC_CLSFLOWER_REPLACE:
 		return i40e_configure_clsflower(vsi, cls_flower);
@@ -7510,6 +7465,9 @@ static int i40e_setup_tc_block_cb(enum tc_setup_type type, void *type_data,
 {
 	struct i40e_netdev_priv *np = cb_priv;
 
+	if (!tc_cls_can_offload_and_chain0(np->vsi->netdev, type_data))
+		return -EOPNOTSUPP;
+
 	switch (type) {
 	case TC_SETUP_CLSFLOWER:
 		return i40e_setup_tc_cls_flower(np, type_data);
@@ -7727,6 +7685,9 @@ static void i40e_fdir_filter_exit(struct i40e_pf *pf)
 	/* Reprogram the default input set for Other/IPv4 */
 	i40e_write_fd_input_set(pf, I40E_FILTER_PCTYPE_NONF_IPV4_OTHER,
 				I40E_L3_SRC_MASK | I40E_L3_DST_MASK);
+
+	i40e_write_fd_input_set(pf, I40E_FILTER_PCTYPE_FRAG_IPV4,
+				I40E_L3_SRC_MASK | I40E_L3_DST_MASK);
 }
 
 /**
@@ -9060,6 +9021,17 @@ static int i40e_rebuild_channels(struct i40e_vsi *vsi)
 				 vsi->uplink_seid);
 			return ret;
 		}
+		/* Reconfigure TX queues using QTX_CTL register */
+		ret = i40e_channel_config_tx_ring(vsi->back, vsi, ch);
+		if (ret) {
+			dev_info(&vsi->back->pdev->dev,
+				 "failed to configure TX rings for channel %u\n",
+				 ch->seid);
+			return ret;
+		}
+		/* update 'next_base_queue' */
+		vsi->next_base_queue = vsi->next_base_queue +
+							ch->num_queue_pairs;
 		if (ch->max_tx_rate) {
 			u64 credits = ch->max_tx_rate;
 
@@ -9263,6 +9235,9 @@ static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired)
 		goto end_core_reset;
 	}
 
+	/* Enable FW to write a default DCB config on link-up */
+	i40e_aq_set_dcb_parameters(hw, true, NULL);
+
 #ifdef CONFIG_I40E_DCB
 	ret = i40e_init_pf_dcb(pf);
 	if (ret) {
@@ -9680,7 +9655,7 @@ static void i40e_service_task(struct work_struct *work)
 	if (test_and_set_bit(__I40E_SERVICE_SCHED, pf->state))
 		return;
 
-	i40e_detect_recover_hung(pf);
+	i40e_detect_recover_hung(pf->vsi[pf->lan_vsi]);
 	i40e_sync_filters_subtask(pf);
 	i40e_reset_subtask(pf);
 	i40e_handle_mdd_event(pf);
@@ -10447,10 +10422,9 @@ static int i40e_init_interrupt_scheme(struct i40e_pf *pf)
 	/* set up vector assignment tracking */
 	size = sizeof(struct i40e_lump_tracking) + (sizeof(u16) * vectors);
 	pf->irq_pile = kzalloc(size, GFP_KERNEL);
-	if (!pf->irq_pile) {
-		dev_err(&pf->pdev->dev, "error allocating irq_pile memory\n");
+	if (!pf->irq_pile)
 		return -ENOMEM;
-	}
+
 	pf->irq_pile->num_entries = vectors;
 	pf->irq_pile->search_hint = 0;
 
@@ -10768,8 +10742,13 @@ static int i40e_pf_config_rss(struct i40e_pf *pf)
 	/* Determine the RSS size of the VSI */
 	if (!vsi->rss_size) {
 		u16 qcount;
-
-		qcount = vsi->num_queue_pairs / vsi->tc_config.numtc;
+		/* If the firmware does something weird during VSI init, we
+		 * could end up with zero TCs. Check for that to avoid
+		 * divide-by-zero. It probably won't pass traffic, but it also
+		 * won't panic.
+		 */
+		qcount = vsi->num_queue_pairs /
+			 (vsi->tc_config.numtc ? vsi->tc_config.numtc : 1);
 		vsi->rss_size = min_t(int, pf->alloc_rss_size, qcount);
 	}
 	if (!vsi->rss_size)
@@ -10957,7 +10936,7 @@ i40e_status i40e_commit_partition_bw_setting(struct i40e_pf *pf)
 	ret = i40e_aq_update_nvm(&pf->hw,
 				 I40E_SR_NVM_CONTROL_WORD,
 				 0x10, sizeof(nvm_word),
-				 &nvm_word, true, NULL);
+				 &nvm_word, true, 0, NULL);
 	/* Save off last admin queue command status before releasing
 	 * the NVM
 	 */
@@ -11098,13 +11077,13 @@ static int i40e_sw_init(struct i40e_pf *pf)
 	    pf->hw.aq.fw_maj_ver >= 6)
 		pf->hw_features |= I40E_HW_PTP_L4_CAPABLE;
 
-	if (pf->hw.func_caps.vmdq) {
+	if (pf->hw.func_caps.vmdq && num_online_cpus() != 1) {
 		pf->num_vmdq_vsis = I40E_DEFAULT_NUM_VMDQ_VSI;
 		pf->flags |= I40E_FLAG_VMDQ_ENABLED;
 		pf->num_vmdq_qps = i40e_default_queues_per_vmdq(pf);
 	}
 
-	if (pf->hw.func_caps.iwarp) {
+	if (pf->hw.func_caps.iwarp && num_online_cpus() != 1) {
 		pf->flags |= I40E_FLAG_IWARP_ENABLED;
 		/* IWARP needs one extra vector for CQP just like MISC.*/
 		pf->num_iwarp_msix = (int)num_online_cpus() + 1;
@@ -13581,6 +13560,10 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_drvdata(pdev, pf);
 	pci_save_state(pdev);
+
+	/* Enable FW to write default DCB config on link-up */
+	i40e_aq_set_dcb_parameters(hw, true, NULL);
+
 #ifdef CONFIG_I40E_DCB
 	err = i40e_init_pf_dcb(pf);
 	if (err) {