Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue

Jeff Kirsher says:

====================
100GbE Intel Wired LAN Driver Updates 2017-10-03

This series contains updates to fm10k only.

Jake provides majority of the changes in this series, starting with using
fm10k_prepare_for_reset() if we lose PCIe link.  Before we would detach
the device and close the netdev, which left a lot of items still active,
such as the Tx/Rx resources.  This could cause problems where register
reads would return potentially invalid values and would result in unknown
driver behavior, so call fm10k_prepare_for_reset() much like we do for
suspend/resume cycles.  This will attempt to shutdown as much as possible
to prevent possible issues.  Then replaced the PCI specific legacy power
management hooks with the new generic power management hooks for both
suspend and hibernate.  Introduced a workqueue item which monitors a
queue of MAC and VLAN requests since a large number of MAC address or
VLAN updates at once can overload the mailbox with too many messages at
once.  Fixed a cppcheck warning by properly declaring the min_rate and
max_rate variables in the declaration and definition for .ndo_set_vf_bw,
rather than using "unused" for the minimum rates.

Joe Perches fixes the backward logic when using net_ratelimit().
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/intel/fm10k/fm10k.h

@@ -248,6 +248,29 @@ struct fm10k_udp_port {
 	__be16			port;
 };
 
+enum fm10k_macvlan_request_type {
+	FM10K_UC_MAC_REQUEST,
+	FM10K_MC_MAC_REQUEST,
+	FM10K_VLAN_REQUEST
+};
+
+struct fm10k_macvlan_request {
+	enum fm10k_macvlan_request_type type;
+	struct list_head list;
+	union {
+		struct fm10k_mac_request {
+			u8 addr[ETH_ALEN];
+			u16 glort;
+			u16 vid;
+		} mac;
+		struct fm10k_vlan_request {
+			u32 vid;
+			u8 vsi;
+		} vlan;
+	};
+	bool set;
+};
+
 /* one work queue for entire driver */
 extern struct workqueue_struct *fm10k_workqueue;
 
@@ -270,11 +293,15 @@ enum fm10k_flags_t {
 
 enum fm10k_state_t {
 	__FM10K_RESETTING,
+	__FM10K_RESET_DETACHED,
+	__FM10K_RESET_SUSPENDED,
 	__FM10K_DOWN,
 	__FM10K_SERVICE_SCHED,
 	__FM10K_SERVICE_REQUEST,
 	__FM10K_SERVICE_DISABLE,
-	__FM10K_MBX_LOCK,
+	__FM10K_MACVLAN_SCHED,
+	__FM10K_MACVLAN_REQUEST,
+	__FM10K_MACVLAN_DISABLE,
 	__FM10K_LINK_DOWN,
 	__FM10K_UPDATING_STATS,
 	/* This value must be last and determines the BITMAP size */
@@ -344,6 +371,8 @@ struct fm10k_intfc {
 
 	struct fm10k_hw_stats stats;
 	struct fm10k_hw hw;
+	/* Mailbox lock */
+	spinlock_t mbx_lock;
 	u32 __iomem *uc_addr;
 	u32 __iomem *sw_addr;
 	u16 msg_enable;
@@ -365,6 +394,12 @@ struct fm10k_intfc {
 	struct list_head vxlan_port;
 	struct list_head geneve_port;
 
+	/* MAC/VLAN update queue */
+	struct list_head macvlan_requests;
+	struct delayed_work macvlan_task;
+	/* MAC/VLAN update queue lock */
+	spinlock_t macvlan_lock;
+
 #ifdef CONFIG_DEBUG_FS
 	struct dentry *dbg_intfc;
 #endif /* CONFIG_DEBUG_FS */
@@ -384,23 +419,17 @@ struct fm10k_intfc {
 
 static inline void fm10k_mbx_lock(struct fm10k_intfc *interface)
 {
-	/* busy loop if we cannot obtain the lock as some calls
-	 * such as ndo_set_rx_mode may be made in atomic context
-	 */
-	while (test_and_set_bit(__FM10K_MBX_LOCK, interface->state))
-		udelay(20);
+	spin_lock(&interface->mbx_lock);
 }
 
 static inline void fm10k_mbx_unlock(struct fm10k_intfc *interface)
 {
-	/* flush memory to make sure state is correct */
-	smp_mb__before_atomic();
-	clear_bit(__FM10K_MBX_LOCK, interface->state);
+	spin_unlock(&interface->mbx_lock);
 }
 
 static inline int fm10k_mbx_trylock(struct fm10k_intfc *interface)
 {
-	return !test_and_set_bit(__FM10K_MBX_LOCK, interface->state);
+	return spin_trylock(&interface->mbx_lock);
 }
 
 /* fm10k_test_staterr - test bits in Rx descriptor status and error fields */
@@ -490,6 +519,7 @@ void fm10k_up(struct fm10k_intfc *interface);
 void fm10k_down(struct fm10k_intfc *interface);
 void fm10k_update_stats(struct fm10k_intfc *interface);
 void fm10k_service_event_schedule(struct fm10k_intfc *interface);
+void fm10k_macvlan_schedule(struct fm10k_intfc *interface);
 void fm10k_update_rx_drop_en(struct fm10k_intfc *interface);
 #ifdef CONFIG_NET_POLL_CONTROLLER
 void fm10k_netpoll(struct net_device *netdev);
@@ -510,6 +540,12 @@ void fm10k_reset_rx_state(struct fm10k_intfc *);
 int fm10k_setup_tc(struct net_device *dev, u8 tc);
 int fm10k_open(struct net_device *netdev);
 int fm10k_close(struct net_device *netdev);
+int fm10k_queue_vlan_request(struct fm10k_intfc *interface, u32 vid,
+			     u8 vsi, bool set);
+int fm10k_queue_mac_request(struct fm10k_intfc *interface, u16 glort,
+			    const unsigned char *addr, u16 vid, bool set);
+void fm10k_clear_macvlan_queue(struct fm10k_intfc *interface,
+			       u16 glort, bool vlans);
 
 /* Ethtool */
 void fm10k_set_ethtool_ops(struct net_device *dev);
@@ -526,8 +562,8 @@ s32 fm10k_iov_update_pvid(struct fm10k_intfc *interface, u16 glort, u16 pvid);
 int fm10k_ndo_set_vf_mac(struct net_device *netdev, int vf_idx, u8 *mac);
 int fm10k_ndo_set_vf_vlan(struct net_device *netdev,
 			  int vf_idx, u16 vid, u8 qos, __be16 vlan_proto);
-int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx, int rate,
-			int unused);
+int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
+			int __always_unused min_rate, int max_rate);
 int fm10k_ndo_get_vf_config(struct net_device *netdev,
 			    int vf_idx, struct ifla_vf_info *ivi);
 

drivers/net/ethernet/intel/fm10k/fm10k_iov.c

@@ -35,10 +35,133 @@ static s32 fm10k_iov_msg_error(struct fm10k_hw *hw, u32 **results,
 	return fm10k_tlv_msg_error(hw, results, mbx);
 }
 
+/**
+ *  fm10k_iov_msg_queue_mac_vlan - Message handler for MAC/VLAN request from VF
+ *  @hw: Pointer to hardware structure
+ *  @results: Pointer array to message, results[0] is pointer to message
+ *  @mbx: Pointer to mailbox information structure
+ *
+ *  This function is a custom handler for MAC/VLAN requests from the VF. The
+ *  assumption is that it is acceptable to directly hand off the message from
+ *  the VF to the PF's switch manager. However, we use a MAC/VLAN message
+ *  queue to avoid overloading the mailbox when a large number of requests
+ *  come in.
+ **/
+static s32 fm10k_iov_msg_queue_mac_vlan(struct fm10k_hw *hw, u32 **results,
+					struct fm10k_mbx_info *mbx)
+{
+	struct fm10k_vf_info *vf_info = (struct fm10k_vf_info *)mbx;
+	struct fm10k_intfc *interface = hw->back;
+	u8 mac[ETH_ALEN];
+	u32 *result;
+	int err = 0;
+	bool set;
+	u16 vlan;
+	u32 vid;
+
+	/* we shouldn't be updating rules on a disabled interface */
+	if (!FM10K_VF_FLAG_ENABLED(vf_info))
+		err = FM10K_ERR_PARAM;
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_VLAN]) {
+		result = results[FM10K_MAC_VLAN_MSG_VLAN];
+
+		/* record VLAN id requested */
+		err = fm10k_tlv_attr_get_u32(result, &vid);
+		if (err)
+			return err;
+
+		set = !(vid & FM10K_VLAN_CLEAR);
+		vid &= ~FM10K_VLAN_CLEAR;
+
+		/* if the length field has been set, this is a multi-bit
+		 * update request. For multi-bit requests, simply disallow
+		 * them when the pf_vid has been set. In this case, the PF
+		 * should have already cleared the VLAN_TABLE, and if we
+		 * allowed them, it could allow a rogue VF to receive traffic
+		 * on a VLAN it was not assigned. In the single-bit case, we
+		 * need to modify requests for VLAN 0 to use the default PF or
+		 * SW vid when assigned.
+		 */
+
+		if (vid >> 16) {
+			/* prevent multi-bit requests when PF has
+			 * administratively set the VLAN for this VF
+			 */
+			if (vf_info->pf_vid)
+				return FM10K_ERR_PARAM;
+		} else {
+			err = fm10k_iov_select_vid(vf_info, (u16)vid);
+			if (err < 0)
+				return err;
+
+			vid = err;
+		}
+
+		/* update VSI info for VF in regards to VLAN table */
+		err = hw->mac.ops.update_vlan(hw, vid, vf_info->vsi, set);
+	}
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_MAC]) {
+		result = results[FM10K_MAC_VLAN_MSG_MAC];
+
+		/* record unicast MAC address requested */
+		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
+		if (err)
+			return err;
+
+		/* block attempts to set MAC for a locked device */
+		if (is_valid_ether_addr(vf_info->mac) &&
+		    !ether_addr_equal(mac, vf_info->mac))
+			return FM10K_ERR_PARAM;
+
+		set = !(vlan & FM10K_VLAN_CLEAR);
+		vlan &= ~FM10K_VLAN_CLEAR;
+
+		err = fm10k_iov_select_vid(vf_info, vlan);
+		if (err < 0)
+			return err;
+
+		vlan = (u16)err;
+
+		/* Add this request to the MAC/VLAN queue */
+		err = fm10k_queue_mac_request(interface, vf_info->glort,
+					      mac, vlan, set);
+	}
+
+	if (!err && !!results[FM10K_MAC_VLAN_MSG_MULTICAST]) {
+		result = results[FM10K_MAC_VLAN_MSG_MULTICAST];
+
+		/* record multicast MAC address requested */
+		err = fm10k_tlv_attr_get_mac_vlan(result, mac, &vlan);
+		if (err)
+			return err;
+
+		/* verify that the VF is allowed to request multicast */
+		if (!(vf_info->vf_flags & FM10K_VF_FLAG_MULTI_ENABLED))
+			return FM10K_ERR_PARAM;
+
+		set = !(vlan & FM10K_VLAN_CLEAR);
+		vlan &= ~FM10K_VLAN_CLEAR;
+
+		err = fm10k_iov_select_vid(vf_info, vlan);
+		if (err < 0)
+			return err;
+
+		vlan = (u16)err;
+
+		/* Add this request to the MAC/VLAN queue */
+		err = fm10k_queue_mac_request(interface, vf_info->glort,
+					      mac, vlan, set);
+	}
+
+	return err;
+}
+
 static const struct fm10k_msg_data iov_mbx_data[] = {
 	FM10K_TLV_MSG_TEST_HANDLER(fm10k_tlv_msg_test),
 	FM10K_VF_MSG_MSIX_HANDLER(fm10k_iov_msg_msix_pf),
-	FM10K_VF_MSG_MAC_VLAN_HANDLER(fm10k_iov_msg_mac_vlan_pf),
+	FM10K_VF_MSG_MAC_VLAN_HANDLER(fm10k_iov_msg_queue_mac_vlan),
 	FM10K_VF_MSG_LPORT_STATE_HANDLER(fm10k_iov_msg_lport_state_pf),
 	FM10K_TLV_MSG_ERROR_HANDLER(fm10k_iov_msg_error),
 };
@@ -126,8 +249,10 @@ process_mbx:
 		hw->mbx.ops.process(hw, &hw->mbx);
 
 		/* verify port mapping is valid, if not reset port */
-		if (vf_info->vf_flags && !fm10k_glort_valid_pf(hw, glort))
+		if (vf_info->vf_flags && !fm10k_glort_valid_pf(hw, glort)) {
 			hw->iov.ops.reset_lport(hw, vf_info);
+			fm10k_clear_macvlan_queue(interface, glort, false);
+		}
 
 		/* reset VFs that have mailbox timed out */
 		if (!mbx->timeout) {
@@ -190,6 +315,7 @@ void fm10k_iov_suspend(struct pci_dev *pdev)
 
 		hw->iov.ops.reset_resources(hw, vf_info);
 		hw->iov.ops.reset_lport(hw, vf_info);
+		fm10k_clear_macvlan_queue(interface, vf_info->glort, false);
 	}
 }
 
@@ -414,6 +540,8 @@ static inline void fm10k_reset_vf_info(struct fm10k_intfc *interface,
 	/* disable LPORT for this VF which clears switch rules */
 	hw->iov.ops.reset_lport(hw, vf_info);
 
+	fm10k_clear_macvlan_queue(interface, vf_info->glort, false);
+
 	/* assign new MAC+VLAN for this VF */
 	hw->iov.ops.assign_default_mac_vlan(hw, vf_info);
 
@@ -485,7 +613,7 @@ int fm10k_ndo_set_vf_vlan(struct net_device *netdev, int vf_idx, u16 vid,
 }
 
 int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
-			int __always_unused unused, int rate)
+			int __always_unused min_rate, int max_rate)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
 	struct fm10k_iov_data *iov_data = interface->iov_data;
@@ -496,14 +624,15 @@ int fm10k_ndo_set_vf_bw(struct net_device *netdev, int vf_idx,
 		return -EINVAL;
 
 	/* rate limit cannot be less than 10Mbs or greater than link speed */
-	if (rate && ((rate < FM10K_VF_TC_MIN) || rate > FM10K_VF_TC_MAX))
+	if (max_rate &&
+	    (max_rate < FM10K_VF_TC_MIN || max_rate > FM10K_VF_TC_MAX))
 		return -EINVAL;
 
 	/* store values */
-	iov_data->vf_info[vf_idx].rate = rate;
+	iov_data->vf_info[vf_idx].rate = max_rate;
 
 	/* update hardware configuration */
-	hw->iov.ops.configure_tc(hw, vf_idx, rate);
+	hw->iov.ops.configure_tc(hw, vf_idx, max_rate);
 
 	return 0;
 }

drivers/net/ethernet/intel/fm10k/fm10k_main.c

@@ -28,7 +28,7 @@
 
 #include "fm10k.h"
 
-#define DRV_VERSION	"0.21.7-k"
+#define DRV_VERSION	"0.22.1-k"
 #define DRV_SUMMARY	"Intel(R) Ethernet Switch Host Interface Driver"
 const char fm10k_driver_version[] = DRV_VERSION;
 char fm10k_driver_name[] = "fm10k";
@@ -806,9 +806,10 @@ static int fm10k_tso(struct fm10k_ring *tx_ring,
 	tx_desc->mss = cpu_to_le16(skb_shinfo(skb)->gso_size);
 
 	return 1;
+
 err_vxlan:
 	tx_ring->netdev->features &= ~NETIF_F_GSO_UDP_TUNNEL;
-	if (!net_ratelimit())
+	if (net_ratelimit())
 		netdev_err(tx_ring->netdev,
 			   "TSO requested for unsupported tunnel, disabling offload\n");
 	return -1;

drivers/net/ethernet/intel/fm10k/fm10k_netdev.c

@@ -758,11 +758,132 @@ static bool fm10k_host_mbx_ready(struct fm10k_intfc *interface)
 	return (hw->mac.type == fm10k_mac_vf || interface->host_ready);
 }
 
+/**
+ * fm10k_queue_vlan_request - Queue a VLAN update request
+ * @interface: the fm10k interface structure
+ * @vid: the VLAN vid
+ * @vsi: VSI index number
+ * @set: whether to set or clear
+ *
+ * This function queues up a VLAN update. For VFs, this must be sent to the
+ * managing PF over the mailbox. For PFs, we'll use the same handling so that
+ * it's similar to the VF. This avoids storming the PF<->VF mailbox with too
+ * many VLAN updates during reset.
+ */
+int fm10k_queue_vlan_request(struct fm10k_intfc *interface,
+			     u32 vid, u8 vsi, bool set)
+{
+	struct fm10k_macvlan_request *request;
+	unsigned long flags;
+
+	/* This must be atomic since we may be called while the netdev
+	 * addr_list_lock is held
+	 */
+	request = kzalloc(sizeof(*request), GFP_ATOMIC);
+	if (!request)
+		return -ENOMEM;
+
+	request->type = FM10K_VLAN_REQUEST;
+	request->vlan.vid = vid;
+	request->vlan.vsi = vsi;
+	request->set = set;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+	list_add_tail(&request->list, &interface->macvlan_requests);
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+	fm10k_macvlan_schedule(interface);
+
+	return 0;
+}
+
+/**
+ * fm10k_queue_mac_request - Queue a MAC update request
+ * @interface: the fm10k interface structure
+ * @glort: the target glort for this update
+ * @addr: the address to update
+ * @vid: the vid to update
+ * @sync: whether to add or remove
+ *
+ * This function queues up a MAC request for sending to the switch manager.
+ * A separate thread monitors the queue and sends updates to the switch
+ * manager. Return 0 on success, and negative error code on failure.
+ **/
+int fm10k_queue_mac_request(struct fm10k_intfc *interface, u16 glort,
+			    const unsigned char *addr, u16 vid, bool set)
+{
+	struct fm10k_macvlan_request *request;
+	unsigned long flags;
+
+	/* This must be atomic since we may be called while the netdev
+	 * addr_list_lock is held
+	 */
+	request = kzalloc(sizeof(*request), GFP_ATOMIC);
+	if (!request)
+		return -ENOMEM;
+
+	if (is_multicast_ether_addr(addr))
+		request->type = FM10K_MC_MAC_REQUEST;
+	else
+		request->type = FM10K_UC_MAC_REQUEST;
+
+	ether_addr_copy(request->mac.addr, addr);
+	request->mac.glort = glort;
+	request->mac.vid = vid;
+	request->set = set;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+	list_add_tail(&request->list, &interface->macvlan_requests);
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+	fm10k_macvlan_schedule(interface);
+
+	return 0;
+}
+
+/**
+ * fm10k_clear_macvlan_queue - Cancel pending updates for a given glort
+ * @interface: the fm10k interface structure
+ * @glort: the target glort to clear
+ * @vlans: true to clear VLAN messages, false to ignore them
+ *
+ * Cancel any outstanding MAC/VLAN requests for a given glort. This is
+ * expected to be called when a logical port goes down.
+ **/
+void fm10k_clear_macvlan_queue(struct fm10k_intfc *interface,
+			       u16 glort, bool vlans)
+
+{
+	struct fm10k_macvlan_request *r, *tmp;
+	unsigned long flags;
+
+	spin_lock_irqsave(&interface->macvlan_lock, flags);
+
+	/* Free any outstanding MAC/VLAN requests for this interface */
+	list_for_each_entry_safe(r, tmp, &interface->macvlan_requests, list) {
+		switch (r->type) {
+		case FM10K_MC_MAC_REQUEST:
+		case FM10K_UC_MAC_REQUEST:
+			/* Don't free requests for other interfaces */
+			if (r->mac.glort != glort)
+				break;
+			/* fall through */
+		case FM10K_VLAN_REQUEST:
+			if (vlans) {
+				list_del(&r->list);
+				kfree(r);
+			}
+			break;
+		}
+	}
+
+	spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+}
+
 static int fm10k_uc_vlan_unsync(struct net_device *netdev,
 				const unsigned char *uc_addr)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 glort = interface->glort;
 	u16 vid = interface->vid;
 	bool set = !!(vid / VLAN_N_VID);
@@ -771,10 +892,7 @@ static int fm10k_uc_vlan_unsync(struct net_device *netdev,
 	/* drop any leading bits on the VLAN ID */
 	vid &= VLAN_N_VID - 1;
 
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_uc_addr(hw, glort, uc_addr,
-						 vid, set, 0);
-
+	err = fm10k_queue_mac_request(interface, glort, uc_addr, vid, set);
 	if (err)
 		return err;
 
@@ -786,7 +904,6 @@ static int fm10k_mc_vlan_unsync(struct net_device *netdev,
 				const unsigned char *mc_addr)
 {
 	struct fm10k_intfc *interface = netdev_priv(netdev);
-	struct fm10k_hw *hw = &interface->hw;
 	u16 glort = interface->glort;
 	u16 vid = interface->vid;
 	bool set = !!(vid / VLAN_N_VID);
@@ -795,9 +912,7 @@ static int fm10k_mc_vlan_unsync(struct net_device *netdev,
 	/* drop any leading bits on the VLAN ID */
 	vid &= VLAN_N_VID - 1;
 
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_mc_addr(hw, glort, mc_addr, vid, set);
-
+	err = fm10k_queue_mac_request(interface, glort, mc_addr, vid, set);
 	if (err)
 		return err;
 
@@ -855,18 +970,14 @@ static int fm10k_update_vid(struct net_device *netdev, u16 vid, bool set)
 
 	/* only need to update the VLAN if not in promiscuous mode */
 	if (!(netdev->flags & IFF_PROMISC)) {
-		err = hw->mac.ops.update_vlan(hw, vid, 0, set);
+		err = fm10k_queue_vlan_request(interface, vid, 0, set);
 		if (err)
 			goto err_out;
 	}
 
-	/* update our base MAC address if host's mailbox is ready */
-	if (fm10k_host_mbx_ready(interface))
-		err = hw->mac.ops.update_uc_addr(hw, interface->glort,
-						 hw->mac.addr, vid, set, 0);
-	else
-		err = -EHOSTDOWN;
-
+	/* Update our base MAC address */
+	err = fm10k_queue_mac_request(interface, interface->glort,
+				      hw->mac.addr, vid, set);
 	if (err)
 		goto err_out;
 
@@ -910,7 +1021,6 @@ static u16 fm10k_find_next_vlan(struct fm10k_intfc *interface, u16 vid)
 
 static void fm10k_clear_unused_vlans(struct fm10k_intfc *interface)
 {
-	struct fm10k_hw *hw = &interface->hw;
 	u32 vid, prev_vid;
 
 	/* loop through and find any gaps in the table */
@@ -922,7 +1032,7 @@ static void fm10k_clear_unused_vlans(struct fm10k_intfc *interface)
 
 		/* send request to clear multiple bits at a time */
 		prev_vid += (vid - prev_vid - 1) << FM10K_VLAN_LENGTH_SHIFT;
-		hw->mac.ops.update_vlan(hw, prev_vid, 0, false);
+		fm10k_queue_vlan_request(interface, prev_vid, 0, false);
 	}
 }
 
@@ -937,15 +1047,11 @@ static int __fm10k_uc_sync(struct net_device *dev,
 	if (!is_valid_ether_addr(addr))
 		return -EADDRNOTAVAIL;
 
-	/* update table with current entries if host's mailbox is ready */
-	if (!fm10k_host_mbx_ready(interface))
-		return -EHOSTDOWN;
-
 	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		err = hw->mac.ops.update_uc_addr(hw, glort, addr,
-						 vid, sync, 0);
+		err = fm10k_queue_mac_request(interface, glort,
+					      addr, vid, sync);
 		if (err)
 			return err;
 	}
@@ -1002,15 +1108,18 @@ static int __fm10k_mc_sync(struct net_device *dev,
 	struct fm10k_intfc *interface = netdev_priv(dev);
 	struct fm10k_hw *hw = &interface->hw;
 	u16 vid, glort = interface->glort;
+	s32 err;
 
-	/* update table with current entries if host's mailbox is ready */
-	if (!fm10k_host_mbx_ready(interface))
-		return 0;
+	if (!is_multicast_ether_addr(addr))
+		return -EADDRNOTAVAIL;
 
 	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		hw->mac.ops.update_mc_addr(hw, glort, addr, vid, sync);
+		err = fm10k_queue_mac_request(interface, glort,
+					      addr, vid, sync);
+		if (err)
+			return err;
 	}
 
 	return 0;
@@ -1050,7 +1159,8 @@ static void fm10k_set_rx_mode(struct net_device *dev)
 	if (interface->xcast_mode != xcast_mode) {
 		/* update VLAN table */
 		if (xcast_mode == FM10K_XCAST_MODE_PROMISC)
-			hw->mac.ops.update_vlan(hw, FM10K_VLAN_ALL, 0, true);
+			fm10k_queue_vlan_request(interface, FM10K_VLAN_ALL,
+						 0, true);
 		if (interface->xcast_mode == FM10K_XCAST_MODE_PROMISC)
 			fm10k_clear_unused_vlans(interface);
 
@@ -1098,22 +1208,20 @@ void fm10k_restore_rx_state(struct fm10k_intfc *interface)
 					       interface->glort_count, true);
 
 	/* update VLAN table */
-	hw->mac.ops.update_vlan(hw, FM10K_VLAN_ALL, 0,
-				xcast_mode == FM10K_XCAST_MODE_PROMISC);
+	fm10k_queue_vlan_request(interface, FM10K_VLAN_ALL, 0,
+				 xcast_mode == FM10K_XCAST_MODE_PROMISC);
 
 	/* Add filter for VLAN 0 */
-	hw->mac.ops.update_vlan(hw, 0, 0, true);
+	fm10k_queue_vlan_request(interface, 0, 0, true);
 
 	/* update table with current entries */
 	for (vid = hw->mac.default_vid ? fm10k_find_next_vlan(interface, 0) : 1;
 	     vid < VLAN_N_VID;
 	     vid = fm10k_find_next_vlan(interface, vid)) {
-		hw->mac.ops.update_vlan(hw, vid, 0, true);
+		fm10k_queue_vlan_request(interface, vid, 0, true);
 
-		/* Update unicast entries if host's mailbox is ready */
-		if (fm10k_host_mbx_ready(interface))
-			hw->mac.ops.update_uc_addr(hw, glort, hw->mac.addr,
-						   vid, true, 0);
+		fm10k_queue_mac_request(interface, glort,
+					hw->mac.addr, vid, true);
 	}
 
 	/* update xcast mode before synchronizing addresses if host's mailbox
@@ -1140,6 +1248,13 @@ void fm10k_reset_rx_state(struct fm10k_intfc *interface)
 	struct net_device *netdev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
 
+	/* Wait for MAC/VLAN work to finish */
+	while (test_bit(__FM10K_MACVLAN_SCHED, interface->state))
+		usleep_range(1000, 2000);
+
+	/* Cancel pending MAC/VLAN requests */
+	fm10k_clear_macvlan_queue(interface, interface->glort, true);
+
 	fm10k_mbx_lock(interface);
 
 	/* clear the logical port state on lower device if host's mailbox is
@@ -1374,8 +1489,8 @@ static void *fm10k_dfwd_add_station(struct net_device *dev,
 	if (fm10k_host_mbx_ready(interface)) {
 		hw->mac.ops.update_xcast_mode(hw, glort,
 					      FM10K_XCAST_MODE_MULTI);
-		hw->mac.ops.update_uc_addr(hw, glort, sdev->dev_addr,
-					   0, true, 0);
+		fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+					0, true);
 	}
 
 	fm10k_mbx_unlock(interface);
@@ -1414,8 +1529,8 @@ static void fm10k_dfwd_del_station(struct net_device *dev, void *priv)
 	if (fm10k_host_mbx_ready(interface)) {
 		hw->mac.ops.update_xcast_mode(hw, glort,
 					      FM10K_XCAST_MODE_NONE);
-		hw->mac.ops.update_uc_addr(hw, glort, sdev->dev_addr,
-					   0, false, 0);
+		fm10k_queue_mac_request(interface, glort, sdev->dev_addr,
+					0, false);
 	}
 
 	fm10k_mbx_unlock(interface);

drivers/net/ethernet/intel/fm10k/fm10k_pci.c

@@ -91,6 +91,76 @@ static int fm10k_hw_ready(struct fm10k_intfc *interface)
 	return FM10K_REMOVED(hw->hw_addr) ? -ENODEV : 0;
 }
 
+/**
+ * fm10k_macvlan_schedule - Schedule MAC/VLAN queue task
+ * @interface: fm10k private interface structure
+ *
+ * Schedule the MAC/VLAN queue monitor task. If the MAC/VLAN task cannot be
+ * started immediately, request that it be restarted when possible.
+ */
+void fm10k_macvlan_schedule(struct fm10k_intfc *interface)
+{
+	/* Avoid processing the MAC/VLAN queue when the service task is
+	 * disabled, or when we're resetting the device.
+	 */
+	if (!test_bit(__FM10K_MACVLAN_DISABLE, interface->state) &&
+	    !test_and_set_bit(__FM10K_MACVLAN_SCHED, interface->state)) {
+		clear_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+		/* We delay the actual start of execution in order to allow
+		 * multiple MAC/VLAN updates to accumulate before handling
+		 * them, and to allow some time to let the mailbox drain
+		 * between runs.
+		 */
+		queue_delayed_work(fm10k_workqueue,
+				   &interface->macvlan_task, 10);
+	} else {
+		set_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+	}
+}
+
+/**
+ * fm10k_stop_macvlan_task - Stop the MAC/VLAN queue monitor
+ * @interface: fm10k private interface structure
+ *
+ * Wait until the MAC/VLAN queue task has stopped, and cancel any future
+ * requests.
+ */
+static void fm10k_stop_macvlan_task(struct fm10k_intfc *interface)
+{
+	/* Disable the MAC/VLAN work item */
+	set_bit(__FM10K_MACVLAN_DISABLE, interface->state);
+
+	/* Make sure we waited until any current invocations have stopped */
+	cancel_delayed_work_sync(&interface->macvlan_task);
+
+	/* We set the __FM10K_MACVLAN_SCHED bit when we schedule the task.
+	 * However, it may not be unset of the MAC/VLAN task never actually
+	 * got a chance to run. Since we've canceled the task here, and it
+	 * cannot be rescheuled right now, we need to ensure the scheduled bit
+	 * gets unset.
+	 */
+	clear_bit(__FM10K_MACVLAN_SCHED, interface->state);
+}
+
+/**
+ * fm10k_resume_macvlan_task - Restart the MAC/VLAN queue monitor
+ * @interface: fm10k private interface structure
+ *
+ * Clear the __FM10K_MACVLAN_DISABLE bit and, if a request occurred, schedule
+ * the MAC/VLAN work monitor.
+ */
+static void fm10k_resume_macvlan_task(struct fm10k_intfc *interface)
+{
+	/* Re-enable the MAC/VLAN work item */
+	clear_bit(__FM10K_MACVLAN_DISABLE, interface->state);
+
+	/* We might have received a MAC/VLAN request while disabled. If so,
+	 * kick off the queue now.
+	 */
+	if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state))
+		fm10k_macvlan_schedule(interface);
+}
+
 void fm10k_service_event_schedule(struct fm10k_intfc *interface)
 {
 	if (!test_bit(__FM10K_SERVICE_DISABLE, interface->state) &&
@@ -153,7 +223,15 @@ static void fm10k_service_timer(unsigned long data)
 	fm10k_service_event_schedule(interface);
 }
 
-static void fm10k_prepare_for_reset(struct fm10k_intfc *interface)
+/**
+ * fm10k_prepare_for_reset - Prepare the driver and device for a pending reset
+ * @interface: fm10k private data structure
+ *
+ * This function prepares for a device reset by shutting as much down as we
+ * can. It does nothing and returns false if __FM10K_RESETTING was already set
+ * prior to calling this function. It returns true if it actually did work.
+ */
+static bool fm10k_prepare_for_reset(struct fm10k_intfc *interface)
 {
 	struct net_device *netdev = interface->netdev;
 
@@ -162,8 +240,15 @@ static void fm10k_prepare_for_reset(struct fm10k_intfc *interface)
 	/* put off any impending NetWatchDogTimeout */
 	netif_trans_update(netdev);
 
-	while (test_and_set_bit(__FM10K_RESETTING, interface->state))
-		usleep_range(1000, 2000);
+	/* Nothing to do if a reset is already in progress */
+	if (test_and_set_bit(__FM10K_RESETTING, interface->state))
+		return false;
+
+	/* As the MAC/VLAN task will be accessing registers it must not be
+	 * running while we reset. Although the task will not be scheduled
+	 * once we start resetting it may already be running
+	 */
+	fm10k_stop_macvlan_task(interface);
 
 	rtnl_lock();
 
@@ -181,6 +266,8 @@ static void fm10k_prepare_for_reset(struct fm10k_intfc *interface)
 	interface->last_reset = jiffies + (10 * HZ);
 
 	rtnl_unlock();
+
+	return true;
 }
 
 static int fm10k_handle_reset(struct fm10k_intfc *interface)
@@ -189,6 +276,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 	struct fm10k_hw *hw = &interface->hw;
 	int err;
 
+	WARN_ON(!test_bit(__FM10K_RESETTING, interface->state));
+
 	rtnl_lock();
 
 	pci_set_master(interface->pdev);
@@ -245,6 +334,8 @@ static int fm10k_handle_reset(struct fm10k_intfc *interface)
 
 	rtnl_unlock();
 
+	fm10k_resume_macvlan_task(interface);
+
 	clear_bit(__FM10K_RESETTING, interface->state);
 
 	return err;
@@ -267,51 +358,75 @@ static void fm10k_detach_subtask(struct fm10k_intfc *interface)
 	struct net_device *netdev = interface->netdev;
 	u32 __iomem *hw_addr;
 	u32 value;
+	int err;
 
-	/* do nothing if device is still present or hw_addr is set */
+	/* do nothing if netdev is still present or hw_addr is set */
 	if (netif_device_present(netdev) || interface->hw.hw_addr)
 		return;
 
+	/* We've lost the PCIe register space, and can no longer access the
+	 * device. Shut everything except the detach subtask down and prepare
+	 * to reset the device in case we recover. If we actually prepare for
+	 * reset, indicate that we're detached.
+	 */
+	if (fm10k_prepare_for_reset(interface))
+		set_bit(__FM10K_RESET_DETACHED, interface->state);
+
 	/* check the real address space to see if we've recovered */
 	hw_addr = READ_ONCE(interface->uc_addr);
 	value = readl(hw_addr);
 	if (~value) {
+		/* Make sure the reset was initiated because we detached,
+		 * otherwise we might race with a different reset flow.
+		 */
+		if (!test_and_clear_bit(__FM10K_RESET_DETACHED,
+					interface->state))
+			return;
+
+		/* Restore the hardware address */
 		interface->hw.hw_addr = interface->uc_addr;
+
+		/* PCIe link has been restored, and the device is active
+		 * again. Restore everything and reset the device.
+		 */
+		err = fm10k_handle_reset(interface);
+		if (err) {
+			netdev_err(netdev, "Unable to reset device: %d\n", err);
+			interface->hw.hw_addr = NULL;
+			return;
+		}
+
+		/* Re-attach the netdev */
 		netif_device_attach(netdev);
-		set_bit(FM10K_FLAG_RESET_REQUESTED, interface->flags);
 		netdev_warn(netdev, "PCIe link restored, device now attached\n");
 		return;
 	}
-
-	rtnl_lock();
-
-	if (netif_running(netdev))
-		dev_close(netdev);
-
-	rtnl_unlock();
 }
 
-static void fm10k_reinit(struct fm10k_intfc *interface)
+static void fm10k_reset_subtask(struct fm10k_intfc *interface)
 {
 	int err;
 
-	fm10k_prepare_for_reset(interface);
-
-	err = fm10k_handle_reset(interface);
-	if (err)
-		dev_err(&interface->pdev->dev,
-			"fm10k_handle_reset failed: %d\n", err);
-}
-
-static void fm10k_reset_subtask(struct fm10k_intfc *interface)
-{
 	if (!test_and_clear_bit(FM10K_FLAG_RESET_REQUESTED,
 				interface->flags))
 		return;
 
+	/* If another thread has already prepared to reset the device, we
+	 * should not attempt to handle a reset here, since we'd race with
+	 * that thread. This may happen if we suspend the device or if the
+	 * PCIe link is lost. In this case, we'll just ignore the RESET
+	 * request, as it will (eventually) be taken care of when the thread
+	 * which actually started the reset is finished.
+	 */
+	if (!fm10k_prepare_for_reset(interface))
+		return;
+
 	netdev_err(interface->netdev, "Reset interface\n");
 
-	fm10k_reinit(interface);
+	err = fm10k_handle_reset(interface);
+	if (err)
+		dev_err(&interface->pdev->dev,
+			"fm10k_handle_reset failed: %d\n", err);
 }
 
 /**
@@ -381,6 +496,10 @@ static void fm10k_watchdog_update_host_state(struct fm10k_intfc *interface)
  **/
 static void fm10k_mbx_subtask(struct fm10k_intfc *interface)
 {
+	/* If we're resetting, bail out */
+	if (test_bit(__FM10K_RESETTING, interface->state))
+		return;
+
 	/* process upstream mailbox and update device state */
 	fm10k_watchdog_update_host_state(interface);
 
@@ -630,9 +749,11 @@ static void fm10k_service_task(struct work_struct *work)
 
 	interface = container_of(work, struct fm10k_intfc, service_task);
 
+	/* Check whether we're detached first */
+	fm10k_detach_subtask(interface);
+
 	/* tasks run even when interface is down */
 	fm10k_mbx_subtask(interface);
-	fm10k_detach_subtask(interface);
 	fm10k_reset_subtask(interface);
 
 	/* tasks only run when interface is up */
@@ -643,6 +764,112 @@ static void fm10k_service_task(struct work_struct *work)
 	fm10k_service_event_complete(interface);
 }
 
+/**
+ * fm10k_macvlan_task - send queued MAC/VLAN requests to switch manager
+ * @work: pointer to work_struct containing our data
+ *
+ * This work item handles sending MAC/VLAN updates to the switch manager. When
+ * the interface is up, it will attempt to queue mailbox messages to the
+ * switch manager requesting updates for MAC/VLAN pairs. If the Tx fifo of the
+ * mailbox is full, it will reschedule itself to try again in a short while.
+ * This ensures that the driver does not overload the switch mailbox with too
+ * many simultaneous requests, causing an unnecessary reset.
+ **/
+static void fm10k_macvlan_task(struct work_struct *work)
+{
+	struct fm10k_macvlan_request *item;
+	struct fm10k_intfc *interface;
+	struct delayed_work *dwork;
+	struct list_head *requests;
+	struct fm10k_hw *hw;
+	unsigned long flags;
+
+	dwork = to_delayed_work(work);
+	interface = container_of(dwork, struct fm10k_intfc, macvlan_task);
+	hw = &interface->hw;
+	requests = &interface->macvlan_requests;
+
+	do {
+		/* Pop the first item off the list */
+		spin_lock_irqsave(&interface->macvlan_lock, flags);
+		item = list_first_entry_or_null(requests,
+						struct fm10k_macvlan_request,
+						list);
+		if (item)
+			list_del_init(&item->list);
+
+		spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+
+		/* We have no more items to process */
+		if (!item)
+			goto done;
+
+		fm10k_mbx_lock(interface);
+
+		/* Check that we have plenty of space to send the message. We
+		 * want to ensure that the mailbox stays low enough to avoid a
+		 * change in the host state, otherwise we may see spurious
+		 * link up / link down notifications.
+		 */
+		if (!hw->mbx.ops.tx_ready(&hw->mbx, FM10K_VFMBX_MSG_MTU + 5)) {
+			hw->mbx.ops.process(hw, &hw->mbx);
+			set_bit(__FM10K_MACVLAN_REQUEST, interface->state);
+			fm10k_mbx_unlock(interface);
+
+			/* Put the request back on the list */
+			spin_lock_irqsave(&interface->macvlan_lock, flags);
+			list_add(&item->list, requests);
+			spin_unlock_irqrestore(&interface->macvlan_lock, flags);
+			break;
+		}
+
+		switch (item->type) {
+		case FM10K_MC_MAC_REQUEST:
+			hw->mac.ops.update_mc_addr(hw,
+						   item->mac.glort,
+						   item->mac.addr,
+						   item->mac.vid,
+						   item->set);
+			break;
+		case FM10K_UC_MAC_REQUEST:
+			hw->mac.ops.update_uc_addr(hw,
+						   item->mac.glort,
+						   item->mac.addr,
+						   item->mac.vid,
+						   item->set,
+						   0);
+			break;
+		case FM10K_VLAN_REQUEST:
+			hw->mac.ops.update_vlan(hw,
+						item->vlan.vid,
+						item->vlan.vsi,
+						item->set);
+			break;
+		default:
+			break;
+		}
+
+		fm10k_mbx_unlock(interface);
+
+		/* Free the item now that we've sent the update */
+		kfree(item);
+	} while (true);
+
+done:
+	WARN_ON(!test_bit(__FM10K_MACVLAN_SCHED, interface->state));
+
+	/* flush memory to make sure state is correct */
+	smp_mb__before_atomic();
+	clear_bit(__FM10K_MACVLAN_SCHED, interface->state);
+
+	/* If a MAC/VLAN request was scheduled since we started, we should
+	 * re-schedule. However, there is no reason to re-schedule if there is
+	 * no work to do.
+	 */
+	if (test_bit(__FM10K_MACVLAN_REQUEST, interface->state))
+		fm10k_macvlan_schedule(interface);
+}
+
 /**
  * fm10k_configure_tx_ring - Configure Tx ring after Reset
  * @interface: board private structure
@@ -1875,9 +2102,16 @@ static int fm10k_sw_init(struct fm10k_intfc *interface,
 	INIT_LIST_HEAD(&interface->vxlan_port);
 	INIT_LIST_HEAD(&interface->geneve_port);
 
+	/* Initialize the MAC/VLAN queue */
+	INIT_LIST_HEAD(&interface->macvlan_requests);
+
 	netdev_rss_key_fill(rss_key, sizeof(rss_key));
 	memcpy(interface->rssrk, rss_key, sizeof(rss_key));
 
+	/* Initialize the mailbox lock */
+	spin_lock_init(&interface->mbx_lock);
+	spin_lock_init(&interface->macvlan_lock);
+
 	/* Start off interface as being down */
 	set_bit(__FM10K_DOWN, interface->state);
 	set_bit(__FM10K_UPDATING_STATS, interface->state);
@@ -2085,6 +2319,9 @@ static int fm10k_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		    (unsigned long)interface);
 	INIT_WORK(&interface->service_task, fm10k_service_task);
 
+	/* Setup the MAC/VLAN queue */
+	INIT_DELAYED_WORK(&interface->macvlan_task, fm10k_macvlan_task);
+
 	/* kick off service timer now, even when interface is down */
 	mod_timer(&interface->service_timer, (HZ * 2) + jiffies);
 
@@ -2138,6 +2375,10 @@ static void fm10k_remove(struct pci_dev *pdev)
 	del_timer_sync(&interface->service_timer);
 
 	fm10k_stop_service_event(interface);
+	fm10k_stop_macvlan_task(interface);
+
+	/* Remove all pending MAC/VLAN requests */
+	fm10k_clear_macvlan_queue(interface, interface->glort, true);
 
 	/* free netdev, this may bounce the interrupts due to setup_tc */
 	if (netdev->reg_state == NETREG_REGISTERED)
@@ -2174,10 +2415,14 @@ static void fm10k_prepare_suspend(struct fm10k_intfc *interface)
 	 * a surprise remove if the PCIe device is disabled while we're
 	 * stopped. We stop the watchdog task until after we resume software
 	 * activity.
+	 *
+	 * Note that the MAC/VLAN task will be stopped as part of preparing
+	 * for reset so we don't need to handle it here.
 	 */
 	fm10k_stop_service_event(interface);
 
-	fm10k_prepare_for_reset(interface);
+	if (fm10k_prepare_for_reset(interface))
+		set_bit(__FM10K_RESET_SUSPENDED, interface->state);
 }
 
 static int fm10k_handle_resume(struct fm10k_intfc *interface)
@@ -2185,6 +2430,13 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface)
 	struct fm10k_hw *hw = &interface->hw;
 	int err;
 
+	/* Even if we didn't properly prepare for reset in
+	 * fm10k_prepare_suspend, we'll attempt to resume anyways.
+	 */
+	if (!test_and_clear_bit(__FM10K_RESET_SUSPENDED, interface->state))
+		dev_warn(&interface->pdev->dev,
+			 "Device was shut down as part of suspend... Attempting to recover\n");
+
 	/* reset statistics starting values */
 	hw->mac.ops.rebind_hw_stats(hw, &interface->stats);
 
@@ -2205,41 +2457,27 @@ static int fm10k_handle_resume(struct fm10k_intfc *interface)
 	/* restart the service task */
 	fm10k_start_service_event(interface);
 
+	/* Restart the MAC/VLAN request queue in-case of outstanding events */
+	fm10k_macvlan_schedule(interface);
+
 	return err;
 }
 
 #ifdef CONFIG_PM
 /**
- * fm10k_resume - Restore device to pre-sleep state
- * @pdev: PCI device information struct
+ * fm10k_resume - Generic PM resume hook
+ * @dev: generic device structure
  *
- * fm10k_resume is called after the system has powered back up from a sleep
- * state and is ready to resume operation.  This function is meant to restore
- * the device back to its pre-sleep state.
+ * Generic PM hook used when waking the device from a low power state after
+ * suspend or hibernation. This function does not need to handle lower PCIe
+ * device state as the stack takes care of that for us.
  **/
-static int fm10k_resume(struct pci_dev *pdev)
+static int fm10k_resume(struct device *dev)
 {
-	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
+	struct fm10k_intfc *interface = pci_get_drvdata(to_pci_dev(dev));
 	struct net_device *netdev = interface->netdev;
 	struct fm10k_hw *hw = &interface->hw;
-	u32 err;
-
-	pci_set_power_state(pdev, PCI_D0);
-	pci_restore_state(pdev);
-
-	/* pci_restore_state clears dev->state_saved so call
-	 * pci_save_state to restore it.
-	 */
-	pci_save_state(pdev);
-
-	err = pci_enable_device_mem(pdev);
-	if (err) {
-		dev_err(&pdev->dev, "Cannot enable PCI device from suspend\n");
-		return err;
-	}
-	pci_set_master(pdev);
-
-	pci_wake_from_d3(pdev, false);
+	int err;
 
 	/* refresh hw_addr in case it was dropped */
 	hw->hw_addr = interface->uc_addr;
@@ -2254,36 +2492,27 @@ static int fm10k_resume(struct pci_dev *pdev)
 }
 
 /**
- * fm10k_suspend - Prepare the device for a system sleep state
- * @pdev: PCI device information struct
+ * fm10k_suspend - Generic PM suspend hook
+ * @dev: generic device structure
  *
- * fm10k_suspend is meant to shutdown the device prior to the system entering
- * a sleep state.  The fm10k hardware does not support wake on lan so the
- * driver simply needs to shut down the device so it is in a low power state.
+ * Generic PM hook used when setting the device into a low power state for
+ * system suspend or hibernation. This function does not need to handle lower
+ * PCIe device state as the stack takes care of that for us.
  **/
-static int fm10k_suspend(struct pci_dev *pdev,
-			 pm_message_t __always_unused state)
+static int fm10k_suspend(struct device *dev)
 {
-	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
+	struct fm10k_intfc *interface = pci_get_drvdata(to_pci_dev(dev));
 	struct net_device *netdev = interface->netdev;
-	int err = 0;
 
 	netif_device_detach(netdev);
 
 	fm10k_prepare_suspend(interface);
 
-	err = pci_save_state(pdev);
-	if (err)
-		return err;
-
-	pci_disable_device(pdev);
-	pci_wake_from_d3(pdev, false);
-	pci_set_power_state(pdev, PCI_D3hot);
-
 	return 0;
 }
 
 #endif /* CONFIG_PM */
+
 /**
  * fm10k_io_error_detected - called when PCI error is detected
  * @pdev: Pointer to PCI device
@@ -2359,11 +2588,18 @@ static void fm10k_io_resume(struct pci_dev *pdev)
 
 	if (err)
 		dev_warn(&pdev->dev,
-			 "fm10k_io_resume failed: %d\n", err);
+			 "%s failed: %d\n", __func__, err);
 	else
 		netif_device_attach(netdev);
 }
 
+/**
+ * fm10k_io_reset_prepare - called when PCI function is about to be reset
+ * @pdev: Pointer to PCI device
+ *
+ * This callback is called when the PCI function is about to be reset,
+ * allowing the device driver to prepare for it.
+ */
 static void fm10k_io_reset_prepare(struct pci_dev *pdev)
 {
 	/* warn incase we have any active VF devices */
@@ -2373,6 +2609,13 @@ static void fm10k_io_reset_prepare(struct pci_dev *pdev)
 	fm10k_prepare_suspend(pci_get_drvdata(pdev));
 }
 
+/**
+ * fm10k_io_reset_done - called when PCI function has finished resetting
+ * @pdev: Pointer to PCI device
+ *
+ * This callback is called just after the PCI function is reset, such as via
+ * /sys/class/net/<enpX>/device/reset or similar.
+ */
 static void fm10k_io_reset_done(struct pci_dev *pdev)
 {
 	struct fm10k_intfc *interface = pci_get_drvdata(pdev);
@@ -2380,7 +2623,7 @@ static void fm10k_io_reset_done(struct pci_dev *pdev)
 
 	if (err) {
 		dev_warn(&pdev->dev,
-			 "fm10k_io_reset_notify failed: %d\n", err);
+			 "%s failed: %d\n", __func__, err);
 		netif_device_detach(interface->netdev);
 	}
 }
@@ -2393,15 +2636,18 @@ static const struct pci_error_handlers fm10k_err_handler = {
 	.reset_done = fm10k_io_reset_done,
 };
 
+static SIMPLE_DEV_PM_OPS(fm10k_pm_ops, fm10k_suspend, fm10k_resume);
+
 static struct pci_driver fm10k_driver = {
 	.name			= fm10k_driver_name,
 	.id_table		= fm10k_pci_tbl,
 	.probe			= fm10k_probe,
 	.remove			= fm10k_remove,
 #ifdef CONFIG_PM
-	.suspend		= fm10k_suspend,
-	.resume			= fm10k_resume,
-#endif
+	.driver = {
+		.pm		= &fm10k_pm_ops,
+	},
+#endif /* CONFIG_PM */
 	.sriov_configure	= fm10k_iov_configure,
 	.err_handler		= &fm10k_err_handler
 };

drivers/net/ethernet/intel/fm10k/fm10k_pf.c

@@ -1186,7 +1186,7 @@ s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *hw, u32 **results,
  * Will report an error if the VLAN ID is out of range. For VID = 0, it will
  * return either the pf_vid or sw_vid depending on which one is set.
  */
-static s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid)
+s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid)
 {
 	if (!vid)
 		return vf_info->pf_vid ? vf_info->pf_vid : vf_info->sw_vid;

drivers/net/ethernet/intel/fm10k/fm10k_pf.h

@@ -1,5 +1,5 @@
 /* Intel(R) Ethernet Switch Host Interface Driver
- * Copyright(c) 2013 - 2016 Intel Corporation.
+ * Copyright(c) 2013 - 2017 Intel Corporation.
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms and conditions of the GNU General Public License,
@@ -114,6 +114,7 @@ extern const struct fm10k_tlv_attr fm10k_err_msg_attr[];
 #define FM10K_PF_MSG_ERR_HANDLER(msg, func) \
 	FM10K_MSG_HANDLER(FM10K_PF_MSG_ID_##msg, fm10k_err_msg_attr, func)
 
+s32 fm10k_iov_select_vid(struct fm10k_vf_info *vf_info, u16 vid);
 s32 fm10k_iov_msg_msix_pf(struct fm10k_hw *, u32 **, struct fm10k_mbx_info *);
 s32 fm10k_iov_msg_mac_vlan_pf(struct fm10k_hw *, u32 **,
 			      struct fm10k_mbx_info *);