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

Jeff Kirsher says:

--------------------
This series contains updates to ixgbe and ixgbevf.
 ...
Akeem G. Abodunrin (1):
  igb: reset PHY in the link_up process to recover PHY setting after
    power down.

Alexander Duyck (8):
  ixgbe: Drop probe_vf and merge functionality into ixgbe_enable_sriov
  ixgbe: Change how we check for pre-existing and assigned VFs
  ixgbevf: Add lock around mailbox ops to prevent simultaneous access
  ixgbevf: Add support for PCI error handling
  ixgbe: Fix handling of FDIR_HASH flag
  ixgbe: Reduce Rx header size to what is actually used
  ixgbe: Use num_tcs.pg_tcs as upper limit for TC when checking based
    on UP
  ixgbe: Use 1TC DCB instead of disabling DCB for MSI and legacy
    interrupts

Don Skidmore (1):
  ixgbe: add support for new 82599 device

Greg Rose (1):
  ixgbevf: Fix namespace issue with ixgbe_write_eitr

John Fastabend (2):
  ixgbe: fix RAR entry counting for generic and fdb_add()
  ixgbe: remove extra unused queues in DCB + FCoE case
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/intel/igb/igb_main.c

@@ -1500,11 +1500,12 @@ static void igb_configure(struct igb_adapter *adapter)
  **/
 void igb_power_up_link(struct igb_adapter *adapter)
 {
+	igb_reset_phy(&adapter->hw);
+
 	if (adapter->hw.phy.media_type == e1000_media_type_copper)
 		igb_power_up_phy_copper(&adapter->hw);
 	else
 		igb_power_up_serdes_link_82575(&adapter->hw);
-	igb_reset_phy(&adapter->hw);
 }
 
 /**

drivers/net/ethernet/intel/ixgbe/ixgbe.h

@@ -77,17 +77,18 @@
 #define IXGBE_MAX_FCPAUSE		 0xFFFF
 
 /* Supported Rx Buffer Sizes */
-#define IXGBE_RXBUFFER_512   512    /* Used for packet split */
+#define IXGBE_RXBUFFER_256    256  /* Used for skb receive header */
 #define IXGBE_MAX_RXBUFFER  16384  /* largest size for a single descriptor */
 
 /*
- * NOTE: netdev_alloc_skb reserves up to 64 bytes, NET_IP_ALIGN mans we
- * reserve 2 more, and skb_shared_info adds an additional 384 bytes more,
- * this adds up to 512 bytes of extra data meaning the smallest allocation
- * we could have is 1K.
- * i.e. RXBUFFER_512 --> size-1024 slab
+ * NOTE: netdev_alloc_skb reserves up to 64 bytes, NET_IP_ALIGN means we
+ * reserve 64 more, and skb_shared_info adds an additional 320 bytes more,
+ * this adds up to 448 bytes of extra data.
+ *
+ * Since netdev_alloc_skb now allocates a page fragment we can use a value
+ * of 256 and the resultant skb will have a truesize of 960 or less.
  */
-#define IXGBE_RX_HDR_SIZE IXGBE_RXBUFFER_512
+#define IXGBE_RX_HDR_SIZE IXGBE_RXBUFFER_256
 
 #define MAXIMUM_ETHERNET_VLAN_SIZE (ETH_FRAME_LEN + ETH_FCS_LEN + VLAN_HLEN)
 
@@ -130,7 +131,6 @@ struct vf_data_storage {
 	u16 tx_rate;
 	u16 vlan_count;
 	u8 spoofchk_enabled;
-	struct pci_dev *vfdev;
 };
 
 struct vf_macvlans {

drivers/net/ethernet/intel/ixgbe/ixgbe_dcb.c

@@ -232,18 +232,22 @@ u8 ixgbe_dcb_get_tc_from_up(struct ixgbe_dcb_config *cfg, int direction, u8 up)
 {
 	struct tc_configuration *tc_config = &cfg->tc_config[0];
 	u8 prio_mask = 1 << up;
-	u8 tc;
+	u8 tc = cfg->num_tcs.pg_tcs;
+
+	/* If tc is 0 then DCB is likely not enabled or supported */
+	if (!tc)
+		goto out;
 
 	/*
-	 * Test for TCs 7 through 1 and report the first match we find.  If
+	 * Test from maximum TC to 1 and report the first match we find.  If
 	 * we find no match we can assume that the TC is 0 since the TC must
 	 * be set for all user priorities
 	 */
-	for (tc = MAX_TRAFFIC_CLASS - 1; tc; tc--) {
+	for (tc--; tc; tc--) {
 		if (prio_mask & tc_config[tc].path[direction].up_to_tc_bitmap)
 			break;
 	}
-
+out:
 	return tc;
 }
 

drivers/net/ethernet/intel/ixgbe/ixgbe_lib.c

@@ -370,6 +370,9 @@ static bool ixgbe_set_dcb_sriov_queues(struct ixgbe_adapter *adapter)
 	adapter->ring_feature[RING_F_RSS].indices = 1;
 	adapter->ring_feature[RING_F_RSS].mask = IXGBE_RSS_DISABLED_MASK;
 
+	/* disable ATR as it is not supported when VMDq is enabled */
+	adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
+
 	adapter->num_rx_pools = vmdq_i;
 	adapter->num_rx_queues_per_pool = tcs;
 
@@ -450,6 +453,9 @@ static bool ixgbe_set_dcb_queues(struct ixgbe_adapter *adapter)
 	f->indices = rss_i;
 	f->mask = rss_m;
 
+	/* disable ATR as it is not supported when multiple TCs are enabled */
+	adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
+
 #ifdef IXGBE_FCOE
 	/* FCoE enabled queues require special configuration indexed
 	 * by feature specific indices and offset. Here we map FCoE
@@ -606,16 +612,22 @@ static bool ixgbe_set_rss_queues(struct ixgbe_adapter *adapter)
 	f->indices = rss_i;
 	f->mask = IXGBE_RSS_16Q_MASK;
 
+	/* disable ATR by default, it will be configured below */
+	adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
+
 	/*
 	 * Use Flow Director in addition to RSS to ensure the best
 	 * distribution of flows across cores, even when an FDIR flow
 	 * isn't matched.
 	 */
-	if (adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) {
+	if (rss_i > 1 && adapter->atr_sample_rate) {
 		f = &adapter->ring_feature[RING_F_FDIR];
 
 		f->indices = min_t(u16, num_online_cpus(), f->limit);
 		rss_i = max_t(u16, rss_i, f->indices);
+
+		if (!(adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE))
+			adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
 	}
 
 #ifdef IXGBE_FCOE
@@ -1053,18 +1065,27 @@ static void ixgbe_set_interrupt_capability(struct ixgbe_adapter *adapter)
 			return;
 	}
 
-	adapter->flags &= ~IXGBE_FLAG_DCB_ENABLED;
-	if (adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) {
-		e_err(probe,
-		      "ATR is not supported while multiple "
-		      "queues are disabled.  Disabling Flow Director\n");
+	/* disable DCB if number of TCs exceeds 1 */
+	if (netdev_get_num_tc(adapter->netdev) > 1) {
+		e_err(probe, "num TCs exceeds number of queues - disabling DCB\n");
+		netdev_reset_tc(adapter->netdev);
+
+		if (adapter->hw.mac.type == ixgbe_mac_82598EB)
+			adapter->hw.fc.requested_mode = adapter->last_lfc_mode;
+
+		adapter->flags &= ~IXGBE_FLAG_DCB_ENABLED;
+		adapter->temp_dcb_cfg.pfc_mode_enable = false;
+		adapter->dcb_cfg.pfc_mode_enable = false;
 	}
-	adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
-	adapter->atr_sample_rate = 0;
-	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
-		ixgbe_disable_sriov(adapter);
+	adapter->dcb_cfg.num_tcs.pg_tcs = 1;
+	adapter->dcb_cfg.num_tcs.pfc_tcs = 1;
+
+	/* disable SR-IOV */
+	ixgbe_disable_sriov(adapter);
 
+	/* disable RSS */
 	adapter->ring_feature[RING_F_RSS].limit = 1;
+
 	ixgbe_set_num_queues(adapter);
 	adapter->num_q_vectors = 1;
 

drivers/net/ethernet/intel/ixgbe/ixgbe_main.c

@@ -1517,8 +1517,8 @@ static bool ixgbe_cleanup_headers(struct ixgbe_ring *rx_ring,
 	 * 60 bytes if the skb->len is less than 60 for skb_pad.
 	 */
 	pull_len = skb_frag_size(frag);
-	if (pull_len > 256)
-		pull_len = ixgbe_get_headlen(va, pull_len);
+	if (pull_len > IXGBE_RX_HDR_SIZE)
+		pull_len = ixgbe_get_headlen(va, IXGBE_RX_HDR_SIZE);
 
 	/* align pull length to size of long to optimize memcpy performance */
 	skb_copy_to_linear_data(skb, va, ALIGN(pull_len, sizeof(long)));
@@ -2688,8 +2688,7 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
 		   32;		/* PTHRESH = 32 */
 
 	/* reinitialize flowdirector state */
-	if ((adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) &&
-	    adapter->atr_sample_rate) {
+	if (adapter->flags & IXGBE_FLAG_FDIR_HASH_CAPABLE) {
 		ring->atr_sample_rate = adapter->atr_sample_rate;
 		ring->atr_count = 0;
 		set_bit(__IXGBE_TX_FDIR_INIT_DONE, &ring->state);
@@ -3442,14 +3441,18 @@ static int ixgbe_write_uc_addr_list(struct net_device *netdev)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
-	unsigned int rar_entries = IXGBE_MAX_PF_MACVLANS;
+	unsigned int rar_entries = hw->mac.num_rar_entries - 1;
 	int count = 0;
 
+	/* In SR-IOV mode significantly less RAR entries are available */
+	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
+		rar_entries = IXGBE_MAX_PF_MACVLANS - 1;
+
 	/* return ENOMEM indicating insufficient memory for addresses */
 	if (netdev_uc_count(netdev) > rar_entries)
 		return -ENOMEM;
 
-	if (!netdev_uc_empty(netdev) && rar_entries) {
+	if (!netdev_uc_empty(netdev)) {
 		struct netdev_hw_addr *ha;
 		/* return error if we do not support writing to RAR table */
 		if (!hw->mac.ops.set_rar)
@@ -4419,7 +4422,6 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter)
 		if (hw->device_id == IXGBE_DEV_ID_82599_T3_LOM)
 			adapter->flags2 |= IXGBE_FLAG2_TEMP_SENSOR_CAPABLE;
 		/* Flow Director hash filters enabled */
-		adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
 		adapter->atr_sample_rate = 20;
 		adapter->ring_feature[RING_F_FDIR].limit =
 							 IXGBE_MAX_FDIR_INDICES;
@@ -4490,6 +4492,12 @@ static int __devinit ixgbe_sw_init(struct ixgbe_adapter *adapter)
 	hw->fc.send_xon = true;
 	hw->fc.disable_fc_autoneg = false;
 
+#ifdef CONFIG_PCI_IOV
+	/* assign number of SR-IOV VFs */
+	if (hw->mac.type != ixgbe_mac_82598EB)
+		adapter->num_vfs = (max_vfs > 63) ? 0 : max_vfs;
+
+#endif
 	/* enable itr by default in dynamic mode */
 	adapter->rx_itr_setting = 1;
 	adapter->tx_itr_setting = 1;
@@ -6695,12 +6703,6 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 	struct ixgbe_adapter *adapter = netdev_priv(dev);
 	struct ixgbe_hw *hw = &adapter->hw;
 
-	/* Multiple traffic classes requires multiple queues */
-	if (!(adapter->flags & IXGBE_FLAG_MSIX_ENABLED)) {
-		e_err(drv, "Enable failed, needs MSI-X\n");
-		return -EINVAL;
-	}
-
 	/* Hardware supports up to 8 traffic classes */
 	if (tc > adapter->dcb_cfg.num_tcs.pg_tcs ||
 	    (hw->mac.type == ixgbe_mac_82598EB &&
@@ -6720,7 +6722,6 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 		ixgbe_set_prio_tc_map(adapter);
 
 		adapter->flags |= IXGBE_FLAG_DCB_ENABLED;
-		adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
 
 		if (adapter->hw.mac.type == ixgbe_mac_82598EB) {
 			adapter->last_lfc_mode = adapter->hw.fc.requested_mode;
@@ -6733,7 +6734,6 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
 			adapter->hw.fc.requested_mode = adapter->last_lfc_mode;
 
 		adapter->flags &= ~IXGBE_FLAG_DCB_ENABLED;
-		adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
 
 		adapter->temp_dcb_cfg.pfc_mode_enable = false;
 		adapter->dcb_cfg.pfc_mode_enable = false;
@@ -6802,20 +6802,40 @@ static int ixgbe_set_features(struct net_device *netdev,
 	 * Check if Flow Director n-tuple support was enabled or disabled.  If
 	 * the state changed, we need to reset.
 	 */
-	if (!(features & NETIF_F_NTUPLE)) {
-		if (adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE) {
-			/* turn off Flow Director, set ATR and reset */
-			if (!(adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) &&
-			    !(adapter->flags & IXGBE_FLAG_DCB_ENABLED))
-				adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
-			need_reset = true;
-		}
-		adapter->flags &= ~IXGBE_FLAG_FDIR_PERFECT_CAPABLE;
-	} else if (!(adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)) {
+	switch (features & NETIF_F_NTUPLE) {
+	case NETIF_F_NTUPLE:
 		/* turn off ATR, enable perfect filters and reset */
+		if (!(adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE))
+			need_reset = true;
+
 		adapter->flags &= ~IXGBE_FLAG_FDIR_HASH_CAPABLE;
 		adapter->flags |= IXGBE_FLAG_FDIR_PERFECT_CAPABLE;
-		need_reset = true;
+		break;
+	default:
+		/* turn off perfect filters, enable ATR and reset */
+		if (adapter->flags & IXGBE_FLAG_FDIR_PERFECT_CAPABLE)
+			need_reset = true;
+
+		adapter->flags &= ~IXGBE_FLAG_FDIR_PERFECT_CAPABLE;
+
+		/* We cannot enable ATR if SR-IOV is enabled */
+		if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
+			break;
+
+		/* We cannot enable ATR if we have 2 or more traffic classes */
+		if (netdev_get_num_tc(netdev) > 1)
+			break;
+
+		/* We cannot enable ATR if RSS is disabled */
+		if (adapter->ring_feature[RING_F_RSS].limit <= 1)
+			break;
+
+		/* A sample rate of 0 indicates ATR disabled */
+		if (!adapter->atr_sample_rate)
+			break;
+
+		adapter->flags |= IXGBE_FLAG_FDIR_HASH_CAPABLE;
+		break;
 	}
 
 	if (features & NETIF_F_HW_VLAN_RX)
@@ -6839,7 +6859,10 @@ static int ixgbe_ndo_fdb_add(struct ndmsg *ndm,
 			     u16 flags)
 {
 	struct ixgbe_adapter *adapter = netdev_priv(dev);
-	int err = -EOPNOTSUPP;
+	int err;
+
+	if (!(adapter->flags & IXGBE_FLAG_SRIOV_ENABLED))
+		return -EOPNOTSUPP;
 
 	if (ndm->ndm_state & NUD_PERMANENT) {
 		pr_info("%s: FDB only supports static addresses\n",
@@ -6847,13 +6870,17 @@ static int ixgbe_ndo_fdb_add(struct ndmsg *ndm,
 		return -EINVAL;
 	}
 
-	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
-		if (is_unicast_ether_addr(addr))
+	if (is_unicast_ether_addr(addr)) {
+		u32 rar_uc_entries = IXGBE_MAX_PF_MACVLANS;
+
+		if (netdev_uc_count(dev) < rar_uc_entries)
 			err = dev_uc_add_excl(dev, addr);
-		else if (is_multicast_ether_addr(addr))
-			err = dev_mc_add_excl(dev, addr);
 		else
-			err = -EINVAL;
+			err = -ENOMEM;
+	} else if (is_multicast_ether_addr(addr)) {
+		err = dev_mc_add_excl(dev, addr);
+	} else {
+		err = -EINVAL;
 	}
 
 	/* Only return duplicate errors if NLM_F_EXCL is set */
@@ -6942,26 +6969,6 @@ static const struct net_device_ops ixgbe_netdev_ops = {
 	.ndo_fdb_dump		= ixgbe_ndo_fdb_dump,
 };
 
-static void __devinit ixgbe_probe_vf(struct ixgbe_adapter *adapter,
-				     const struct ixgbe_info *ii)
-{
-#ifdef CONFIG_PCI_IOV
-	struct ixgbe_hw *hw = &adapter->hw;
-
-	if (hw->mac.type == ixgbe_mac_82598EB)
-		return;
-
-	/* The 82599 supports up to 64 VFs per physical function
-	 * but this implementation limits allocation to 63 so that
-	 * basic networking resources are still available to the
-	 * physical function.  If the user requests greater thn
-	 * 63 VFs then it is an error - reset to default of zero.
-	 */
-	adapter->num_vfs = (max_vfs > 63) ? 0 : max_vfs;
-	ixgbe_enable_sriov(adapter, ii);
-#endif /* CONFIG_PCI_IOV */
-}
-
 /**
  * ixgbe_wol_supported - Check whether device supports WoL
  * @hw: hw specific details
@@ -6988,6 +6995,7 @@ int ixgbe_wol_supported(struct ixgbe_adapter *adapter, u16 device_id,
 			if (hw->bus.func != 0)
 				break;
 		case IXGBE_SUBDEV_ID_82599_SFP:
+		case IXGBE_SUBDEV_ID_82599_RNDC:
 			is_wol_supported = 1;
 			break;
 		}
@@ -7035,6 +7043,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	int i, err, pci_using_dac;
 	u8 part_str[IXGBE_PBANUM_LENGTH];
 	unsigned int indices = num_possible_cpus();
+	unsigned int dcb_max = 0;
 #ifdef IXGBE_FCOE
 	u16 device_caps;
 #endif
@@ -7084,15 +7093,16 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	pci_save_state(pdev);
 
 #ifdef CONFIG_IXGBE_DCB
-	indices *= MAX_TRAFFIC_CLASS;
+	if (ii->mac == ixgbe_mac_82598EB)
+		dcb_max = min_t(unsigned int, indices * MAX_TRAFFIC_CLASS,
+				IXGBE_MAX_RSS_INDICES);
+	else
+		dcb_max = min_t(unsigned int, indices * MAX_TRAFFIC_CLASS,
+				IXGBE_MAX_FDIR_INDICES);
 #endif
 
 	if (ii->mac == ixgbe_mac_82598EB)
-#ifdef CONFIG_IXGBE_DCB
-		indices = min_t(unsigned int, indices, MAX_TRAFFIC_CLASS * 4);
-#else
 		indices = min_t(unsigned int, indices, IXGBE_MAX_RSS_INDICES);
-#endif
 	else
 		indices = min_t(unsigned int, indices, IXGBE_MAX_FDIR_INDICES);
 
@@ -7100,6 +7110,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	indices += min_t(unsigned int, num_possible_cpus(),
 			 IXGBE_MAX_FCOE_INDICES);
 #endif
+	indices = max_t(unsigned int, dcb_max, indices);
 	netdev = alloc_etherdev_mq(sizeof(struct ixgbe_adapter), indices);
 	if (!netdev) {
 		err = -ENOMEM;
@@ -7206,8 +7217,10 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 		goto err_sw_init;
 	}
 
-	ixgbe_probe_vf(adapter, ii);
+#ifdef CONFIG_PCI_IOV
+	ixgbe_enable_sriov(adapter, ii);
 
+#endif
 	netdev->features = NETIF_F_SG |
 			   NETIF_F_IP_CSUM |
 			   NETIF_F_IPV6_CSUM |
@@ -7411,8 +7424,7 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 	ixgbe_release_hw_control(adapter);
 	ixgbe_clear_interrupt_scheme(adapter);
 err_sw_init:
-	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED)
-		ixgbe_disable_sriov(adapter);
+	ixgbe_disable_sriov(adapter);
 	adapter->flags2 &= ~IXGBE_FLAG2_SEARCH_FOR_SFP;
 	iounmap(hw->hw_addr);
 err_ioremap:
@@ -7465,13 +7477,7 @@ static void __devexit ixgbe_remove(struct pci_dev *pdev)
 	if (netdev->reg_state == NETREG_REGISTERED)
 		unregister_netdev(netdev);
 
-	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
-		if (!(ixgbe_check_vf_assignment(adapter)))
-			ixgbe_disable_sriov(adapter);
-		else
-			e_dev_warn("Unloading driver while VFs are assigned "
-				   "- VFs will not be deallocated\n");
-	}
+	ixgbe_disable_sriov(adapter);
 
 	ixgbe_clear_interrupt_scheme(adapter);
 

drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c

@@ -44,50 +44,15 @@
 #include "ixgbe_sriov.h"
 
 #ifdef CONFIG_PCI_IOV
-static int ixgbe_find_enabled_vfs(struct ixgbe_adapter *adapter)
-{
-	struct pci_dev *pdev = adapter->pdev;
-	struct pci_dev *pvfdev;
-	u16 vf_devfn = 0;
-	int device_id;
-	int vfs_found = 0;
-
-	switch (adapter->hw.mac.type) {
-	case ixgbe_mac_82599EB:
-		device_id = IXGBE_DEV_ID_82599_VF;
-		break;
-	case ixgbe_mac_X540:
-		device_id = IXGBE_DEV_ID_X540_VF;
-		break;
-	default:
-		device_id = 0;
-		break;
-	}
-
-	vf_devfn = pdev->devfn + 0x80;
-	pvfdev = pci_get_device(PCI_VENDOR_ID_INTEL, device_id, NULL);
-	while (pvfdev) {
-		if (pvfdev->devfn == vf_devfn &&
-		    (pvfdev->bus->number >= pdev->bus->number))
-			vfs_found++;
-		vf_devfn += 2;
-		pvfdev = pci_get_device(PCI_VENDOR_ID_INTEL,
-					device_id, pvfdev);
-	}
-
-	return vfs_found;
-}
-
 void ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
 			 const struct ixgbe_info *ii)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int err = 0;
 	int num_vf_macvlans, i;
 	struct vf_macvlans *mv_list;
 	int pre_existing_vfs = 0;
 
-	pre_existing_vfs = ixgbe_find_enabled_vfs(adapter);
+	pre_existing_vfs = pci_num_vf(adapter->pdev);
 	if (!pre_existing_vfs && !adapter->num_vfs)
 		return;
 
@@ -106,10 +71,21 @@ void ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
 			 "enabled for this device - Please reload all "
 			 "VF drivers to avoid spoofed packet errors\n");
 	} else {
+		int err;
+		/*
+		 * The 82599 supports up to 64 VFs per physical function
+		 * but this implementation limits allocation to 63 so that
+		 * basic networking resources are still available to the
+		 * physical function.  If the user requests greater thn
+		 * 63 VFs then it is an error - reset to default of zero.
+		 */
+		adapter->num_vfs = min_t(unsigned int, adapter->num_vfs, 63);
+
 		err = pci_enable_sriov(adapter->pdev, adapter->num_vfs);
 		if (err) {
 			e_err(probe, "Failed to enable PCI sriov: %d\n", err);
-			goto err_novfs;
+			adapter->num_vfs = 0;
+			return;
 		}
 	}
 
@@ -193,20 +169,48 @@ void ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
 	/* Oh oh */
 	e_err(probe, "Unable to allocate memory for VF Data Storage - "
 	      "SRIOV disabled\n");
-	pci_disable_sriov(adapter->pdev);
+	ixgbe_disable_sriov(adapter);
+}
 
-err_novfs:
-	adapter->flags &= ~IXGBE_FLAG_SRIOV_ENABLED;
-	adapter->num_vfs = 0;
+static bool ixgbe_vfs_are_assigned(struct ixgbe_adapter *adapter)
+{
+	struct pci_dev *pdev = adapter->pdev;
+	struct pci_dev *vfdev;
+	int dev_id;
+
+	switch (adapter->hw.mac.type) {
+	case ixgbe_mac_82599EB:
+		dev_id = IXGBE_DEV_ID_82599_VF;
+		break;
+	case ixgbe_mac_X540:
+		dev_id = IXGBE_DEV_ID_X540_VF;
+		break;
+	default:
+		return false;
+	}
+
+	/* loop through all the VFs to see if we own any that are assigned */
+	vfdev = pci_get_device(PCI_VENDOR_ID_INTEL, dev_id, NULL);
+	while (vfdev) {
+		/* if we don't own it we don't care */
+		if (vfdev->is_virtfn && vfdev->physfn == pdev) {
+			/* if it is assigned we cannot release it */
+			if (vfdev->dev_flags & PCI_DEV_FLAGS_ASSIGNED)
+				return true;
+		}
+
+		vfdev = pci_get_device(PCI_VENDOR_ID_INTEL, dev_id, vfdev);
+	}
+
+	return false;
 }
-#endif /* #ifdef CONFIG_PCI_IOV */
 
+#endif /* #ifdef CONFIG_PCI_IOV */
 void ixgbe_disable_sriov(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	u32 gpie;
 	u32 vmdctl;
-	int i;
 
 	/* set num VFs to 0 to prevent access to vfinfo */
 	adapter->num_vfs = 0;
@@ -219,7 +223,20 @@ void ixgbe_disable_sriov(struct ixgbe_adapter *adapter)
 	kfree(adapter->mv_list);
 	adapter->mv_list = NULL;
 
+	/* if SR-IOV is already disabled then there is nothing to do */
+	if (!(adapter->flags & IXGBE_FLAG_SRIOV_ENABLED))
+		return;
+
 #ifdef CONFIG_PCI_IOV
+	/*
+	 * If our VFs are assigned we cannot shut down SR-IOV
+	 * without causing issues, so just leave the hardware
+	 * available but disabled
+	 */
+	if (ixgbe_vfs_are_assigned(adapter)) {
+		e_dev_warn("Unloading driver while VFs are assigned - VFs will not be deallocated\n");
+		return;
+
 	/* disable iov and allow time for transactions to clear */
 	pci_disable_sriov(adapter->pdev);
 #endif
@@ -244,12 +261,6 @@ void ixgbe_disable_sriov(struct ixgbe_adapter *adapter)
 	/* take a breather then clean up driver data */
 	msleep(100);
 
-	/* Release reference to VF devices */
-	for (i = 0; i < adapter->num_vfs; i++) {
-		if (adapter->vfinfo[i].vfdev)
-			pci_dev_put(adapter->vfinfo[i].vfdev);
-	}
-
 	adapter->flags &= ~IXGBE_FLAG_SRIOV_ENABLED;
 }
 
@@ -483,28 +494,11 @@ static int ixgbe_set_vf_macvlan(struct ixgbe_adapter *adapter,
 	return 0;
 }
 
-int ixgbe_check_vf_assignment(struct ixgbe_adapter *adapter)
-{
-#ifdef CONFIG_PCI_IOV
-	int i;
-	for (i = 0; i < adapter->num_vfs; i++) {
-		if (adapter->vfinfo[i].vfdev->dev_flags &
-				PCI_DEV_FLAGS_ASSIGNED)
-			return true;
-	}
-#endif
-	return false;
-}
-
 int ixgbe_vf_configuration(struct pci_dev *pdev, unsigned int event_mask)
 {
 	unsigned char vf_mac_addr[6];
 	struct ixgbe_adapter *adapter = pci_get_drvdata(pdev);
 	unsigned int vfn = (event_mask & 0x3f);
-	struct pci_dev *pvfdev;
-	unsigned int device_id;
-	u16 thisvf_devfn = (pdev->devfn + 0x80 + (vfn << 1)) |
-				(pdev->devfn & 1);
 
 	bool enable = ((event_mask & 0x10000000U) != 0);
 
@@ -517,31 +511,6 @@ int ixgbe_vf_configuration(struct pci_dev *pdev, unsigned int event_mask)
 		 * for it later.
 		 */
 		memcpy(adapter->vfinfo[vfn].vf_mac_addresses, vf_mac_addr, 6);
-
-		switch (adapter->hw.mac.type) {
-		case ixgbe_mac_82599EB:
-			device_id = IXGBE_DEV_ID_82599_VF;
-			break;
-		case ixgbe_mac_X540:
-			device_id = IXGBE_DEV_ID_X540_VF;
-			break;
-		default:
-			device_id = 0;
-			break;
-		}
-
-		pvfdev = pci_get_device(PCI_VENDOR_ID_INTEL, device_id, NULL);
-		while (pvfdev) {
-			if (pvfdev->devfn == thisvf_devfn)
-				break;
-			pvfdev = pci_get_device(PCI_VENDOR_ID_INTEL,
-						device_id, pvfdev);
-		}
-		if (pvfdev)
-			adapter->vfinfo[vfn].vfdev = pvfdev;
-		else
-			e_err(drv, "Couldn't find pci dev ptr for VF %4.4x\n",
-			      thisvf_devfn);
 	}
 
 	return 0;

drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.h

@@ -42,7 +42,6 @@ int ixgbe_ndo_get_vf_config(struct net_device *netdev,
 			    int vf, struct ifla_vf_info *ivi);
 void ixgbe_check_vf_rate_limit(struct ixgbe_adapter *adapter);
 void ixgbe_disable_sriov(struct ixgbe_adapter *adapter);
-int ixgbe_check_vf_assignment(struct ixgbe_adapter *adapter);
 #ifdef CONFIG_PCI_IOV
 void ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
 			const struct ixgbe_info *ii);

drivers/net/ethernet/intel/ixgbe/ixgbe_type.h

@@ -54,6 +54,7 @@
 #define IXGBE_DEV_ID_82599_BACKPLANE_FCOE       0x152a
 #define IXGBE_DEV_ID_82599_SFP_FCOE      0x1529
 #define IXGBE_SUBDEV_ID_82599_SFP        0x11A9
+#define IXGBE_SUBDEV_ID_82599_RNDC       0x1F72
 #define IXGBE_SUBDEV_ID_82599_560FLR     0x17D0
 #define IXGBE_DEV_ID_82599_SFP_EM        0x1507
 #define IXGBE_DEV_ID_82599_SFP_SF2       0x154D

drivers/net/ethernet/intel/ixgbevf/ixgbevf.h

@@ -249,6 +249,8 @@ struct ixgbevf_adapter {
 	bool link_up;
 
 	struct work_struct watchdog_task;
+
+	spinlock_t mbx_lock;
 };
 
 enum ixbgevf_state_t {
@@ -284,7 +286,6 @@ extern void ixgbevf_free_rx_resources(struct ixgbevf_adapter *,
 extern void ixgbevf_free_tx_resources(struct ixgbevf_adapter *,
 				      struct ixgbevf_ring *);
 extern void ixgbevf_update_stats(struct ixgbevf_adapter *adapter);
-void ixgbevf_write_eitr(struct ixgbevf_q_vector *);
 extern int ethtool_ioctl(struct ifreq *ifr);
 
 extern void ixgbe_napi_add_all(struct ixgbevf_adapter *adapter);

drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c

@@ -540,6 +540,25 @@ static int ixgbevf_poll(struct napi_struct *napi, int budget)
 	return 0;
 }
 
+/**
+ * ixgbevf_write_eitr - write VTEITR register in hardware specific way
+ * @q_vector: structure containing interrupt and ring information
+ */
+static void ixgbevf_write_eitr(struct ixgbevf_q_vector *q_vector)
+{
+	struct ixgbevf_adapter *adapter = q_vector->adapter;
+	struct ixgbe_hw *hw = &adapter->hw;
+	int v_idx = q_vector->v_idx;
+	u32 itr_reg = q_vector->itr & IXGBE_MAX_EITR;
+
+	/*
+	 * set the WDIS bit to not clear the timer bits and cause an
+	 * immediate assertion of the interrupt
+	 */
+	itr_reg |= IXGBE_EITR_CNT_WDIS;
+
+	IXGBE_WRITE_REG(hw, IXGBE_VTEITR(v_idx), itr_reg);
+}
 
 /**
  * ixgbevf_configure_msix - Configure MSI-X hardware
@@ -662,30 +681,6 @@ static void ixgbevf_update_itr(struct ixgbevf_q_vector *q_vector,
 	ring_container->itr = itr_setting;
 }
 
-/**
- * ixgbevf_write_eitr - write VTEITR register in hardware specific way
- * @q_vector: structure containing interrupt and ring information
- *
- * This function is made to be called by ethtool and by the driver
- * when it needs to update VTEITR registers at runtime.  Hardware
- * specific quirks/differences are taken care of here.
- */
-void ixgbevf_write_eitr(struct ixgbevf_q_vector *q_vector)
-{
-	struct ixgbevf_adapter *adapter = q_vector->adapter;
-	struct ixgbe_hw *hw = &adapter->hw;
-	int v_idx = q_vector->v_idx;
-	u32 itr_reg = q_vector->itr & IXGBE_MAX_EITR;
-
-	/*
-	 * set the WDIS bit to not clear the timer bits and cause an
-	 * immediate assertion of the interrupt
-	 */
-	itr_reg |= IXGBE_EITR_CNT_WDIS;
-
-	IXGBE_WRITE_REG(hw, IXGBE_VTEITR(v_idx), itr_reg);
-}
-
 static void ixgbevf_set_itr(struct ixgbevf_q_vector *q_vector)
 {
 	u32 new_itr = q_vector->itr;
@@ -1120,9 +1115,14 @@ static int ixgbevf_vlan_rx_add_vid(struct net_device *netdev, u16 vid)
 	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 
+	spin_lock(&adapter->mbx_lock);
+
 	/* add VID to filter table */
 	if (hw->mac.ops.set_vfta)
 		hw->mac.ops.set_vfta(hw, vid, 0, true);
+
+	spin_unlock(&adapter->mbx_lock);
+
 	set_bit(vid, adapter->active_vlans);
 
 	return 0;
@@ -1133,9 +1133,14 @@ static int ixgbevf_vlan_rx_kill_vid(struct net_device *netdev, u16 vid)
 	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 
+	spin_lock(&adapter->mbx_lock);
+
 	/* remove VID from filter table */
 	if (hw->mac.ops.set_vfta)
 		hw->mac.ops.set_vfta(hw, vid, 0, false);
+
+	spin_unlock(&adapter->mbx_lock);
+
 	clear_bit(vid, adapter->active_vlans);
 
 	return 0;
@@ -1190,11 +1195,15 @@ static void ixgbevf_set_rx_mode(struct net_device *netdev)
 	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
 	struct ixgbe_hw *hw = &adapter->hw;
 
+	spin_lock(&adapter->mbx_lock);
+
 	/* reprogram multicast list */
 	if (hw->mac.ops.update_mc_addr_list)
 		hw->mac.ops.update_mc_addr_list(hw, netdev);
 
 	ixgbevf_write_uc_addr_list(netdev);
+
+	spin_unlock(&adapter->mbx_lock);
 }
 
 static void ixgbevf_napi_enable_all(struct ixgbevf_adapter *adapter)
@@ -1339,6 +1348,8 @@ static void ixgbevf_up_complete(struct ixgbevf_adapter *adapter)
 
 	ixgbevf_configure_msix(adapter);
 
+	spin_lock(&adapter->mbx_lock);
+
 	if (hw->mac.ops.set_rar) {
 		if (is_valid_ether_addr(hw->mac.addr))
 			hw->mac.ops.set_rar(hw, 0, hw->mac.addr, 0);
@@ -1350,6 +1361,8 @@ static void ixgbevf_up_complete(struct ixgbevf_adapter *adapter)
 	msg[1] = netdev->mtu + ETH_HLEN + ETH_FCS_LEN;
 	hw->mbx.ops.write_posted(hw, msg, 2);
 
+	spin_unlock(&adapter->mbx_lock);
+
 	clear_bit(__IXGBEVF_DOWN, &adapter->state);
 	ixgbevf_napi_enable_all(adapter);
 
@@ -1562,11 +1575,15 @@ void ixgbevf_reset(struct ixgbevf_adapter *adapter)
 	struct ixgbe_hw *hw = &adapter->hw;
 	struct net_device *netdev = adapter->netdev;
 
+	spin_lock(&adapter->mbx_lock);
+
 	if (hw->mac.ops.reset_hw(hw))
 		hw_dbg(hw, "PF still resetting\n");
 	else
 		hw->mac.ops.init_hw(hw);
 
+	spin_unlock(&adapter->mbx_lock);
+
 	if (is_valid_ether_addr(adapter->hw.mac.addr)) {
 		memcpy(netdev->dev_addr, adapter->hw.mac.addr,
 		       netdev->addr_len);
@@ -1893,6 +1910,9 @@ static int __devinit ixgbevf_sw_init(struct ixgbevf_adapter *adapter)
 			adapter->netdev->addr_len);
 	}
 
+	/* lock to protect mailbox accesses */
+	spin_lock_init(&adapter->mbx_lock);
+
 	/* Enable dynamic interrupt throttling rates */
 	adapter->rx_itr_setting = 1;
 	adapter->tx_itr_setting = 1;
@@ -2032,8 +2052,16 @@ static void ixgbevf_watchdog_task(struct work_struct *work)
 	 * no LSC interrupt
 	 */
 	if (hw->mac.ops.check_link) {
-		if ((hw->mac.ops.check_link(hw, &link_speed,
-					    &link_up, false)) != 0) {
+		s32 need_reset;
+
+		spin_lock(&adapter->mbx_lock);
+
+		need_reset = hw->mac.ops.check_link(hw, &link_speed,
+						    &link_up, false);
+
+		spin_unlock(&adapter->mbx_lock);
+
+		if (need_reset) {
 			adapter->link_up = link_up;
 			adapter->link_speed = link_speed;
 			netif_carrier_off(netdev);
@@ -2813,9 +2841,13 @@ static int ixgbevf_set_mac(struct net_device *netdev, void *p)
 	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
 	memcpy(hw->mac.addr, addr->sa_data, netdev->addr_len);
 
+	spin_lock(&adapter->mbx_lock);
+
 	if (hw->mac.ops.set_rar)
 		hw->mac.ops.set_rar(hw, 0, hw->mac.addr, 0);
 
+	spin_unlock(&adapter->mbx_lock);
+
 	return 0;
 }
 
@@ -3152,12 +3184,92 @@ static void __devexit ixgbevf_remove(struct pci_dev *pdev)
 	pci_disable_device(pdev);
 }
 
+/**
+ * ixgbevf_io_error_detected - called when PCI error is detected
+ * @pdev: Pointer to PCI device
+ * @state: The current pci connection state
+ *
+ * This function is called after a PCI bus error affecting
+ * this device has been detected.
+ */
+static pci_ers_result_t ixgbevf_io_error_detected(struct pci_dev *pdev,
+						  pci_channel_state_t state)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
+
+	netif_device_detach(netdev);
+
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
+	if (netif_running(netdev))
+		ixgbevf_down(adapter);
+
+	pci_disable_device(pdev);
+
+	/* Request a slot slot reset. */
+	return PCI_ERS_RESULT_NEED_RESET;
+}
+
+/**
+ * ixgbevf_io_slot_reset - called after the pci bus has been reset.
+ * @pdev: Pointer to PCI device
+ *
+ * Restart the card from scratch, as if from a cold-boot. Implementation
+ * resembles the first-half of the ixgbevf_resume routine.
+ */
+static pci_ers_result_t ixgbevf_io_slot_reset(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
+
+	if (pci_enable_device_mem(pdev)) {
+		dev_err(&pdev->dev,
+			"Cannot re-enable PCI device after reset.\n");
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	pci_set_master(pdev);
+
+	ixgbevf_reset(adapter);
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * ixgbevf_io_resume - called when traffic can start flowing again.
+ * @pdev: Pointer to PCI device
+ *
+ * This callback is called when the error recovery driver tells us that
+ * its OK to resume normal operation. Implementation resembles the
+ * second-half of the ixgbevf_resume routine.
+ */
+static void ixgbevf_io_resume(struct pci_dev *pdev)
+{
+	struct net_device *netdev = pci_get_drvdata(pdev);
+	struct ixgbevf_adapter *adapter = netdev_priv(netdev);
+
+	if (netif_running(netdev))
+		ixgbevf_up(adapter);
+
+	netif_device_attach(netdev);
+}
+
+/* PCI Error Recovery (ERS) */
+static struct pci_error_handlers ixgbevf_err_handler = {
+	.error_detected = ixgbevf_io_error_detected,
+	.slot_reset = ixgbevf_io_slot_reset,
+	.resume = ixgbevf_io_resume,
+};
+
 static struct pci_driver ixgbevf_driver = {
 	.name     = ixgbevf_driver_name,
 	.id_table = ixgbevf_pci_tbl,
 	.probe    = ixgbevf_probe,
 	.remove   = __devexit_p(ixgbevf_remove),
 	.shutdown = ixgbevf_shutdown,
+	.err_handler = &ixgbevf_err_handler
 };
 
 /**