1 files changed, 130 insertions, 62 deletions

diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 3ba3c8fb28a5..44142245343d 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -42,9 +42,13 @@
 #define LINKCHANGE_INT (2 * HZ)
 #define VF_TAKEOVER_INT (HZ / 10)
 
+/* Macros to define the context of vf registration */
+#define VF_REG_IN_PROBE		1
+#define VF_REG_IN_NOTIFIER	2
+
 static unsigned int ring_size __ro_after_init = 128;
 module_param(ring_size, uint, 0444);
-MODULE_PARM_DESC(ring_size, "Ring buffer size (# of pages)");
+MODULE_PARM_DESC(ring_size, "Ring buffer size (# of 4K pages)");
 unsigned int netvsc_ring_bytes __ro_after_init;
 
 static const u32 default_msg = NETIF_MSG_DRV | NETIF_MSG_PROBE |
@@ -1229,14 +1233,14 @@ static int netvsc_change_mtu(struct net_device *ndev, int mtu)
 	if (ret)
 		goto rollback_vf;
 
-	ndev->mtu = mtu;
+	WRITE_ONCE(ndev->mtu, mtu);
 
 	ret = netvsc_attach(ndev, device_info);
 	if (!ret)
 		goto out;
 
 	/* Attempt rollback to original MTU */
-	ndev->mtu = orig_mtu;
+	WRITE_ONCE(ndev->mtu, orig_mtu);
 
 	if (netvsc_attach(ndev, device_info))
 		netdev_err(ndev, "restoring mtu failed\n");
@@ -1582,10 +1586,10 @@ static void netvsc_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 	switch (stringset) {
 	case ETH_SS_STATS:
 		for (i = 0; i < ARRAY_SIZE(netvsc_stats); i++)
-			ethtool_sprintf(&p, netvsc_stats[i].name);
+			ethtool_puts(&p, netvsc_stats[i].name);
 
 		for (i = 0; i < ARRAY_SIZE(vf_stats); i++)
-			ethtool_sprintf(&p, vf_stats[i].name);
+			ethtool_puts(&p, vf_stats[i].name);
 
 		for (i = 0; i < nvdev->num_chn; i++) {
 			ethtool_sprintf(&p, "tx_queue_%u_packets", i);
@@ -1752,8 +1756,8 @@ static u32 netvsc_rss_indir_size(struct net_device *dev)
 	return ndc->rx_table_sz;
 }
 
-static int netvsc_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
-			   u8 *hfunc)
+static int netvsc_get_rxfh(struct net_device *dev,
+			   struct ethtool_rxfh_param *rxfh)
 {
 	struct net_device_context *ndc = netdev_priv(dev);
 	struct netvsc_device *ndev = rtnl_dereference(ndc->nvdev);
@@ -1763,47 +1767,49 @@ static int netvsc_get_rxfh(struct net_device *dev, u32 *indir, u8 *key,
 	if (!ndev)
 		return -ENODEV;
 
-	if (hfunc)
-		*hfunc = ETH_RSS_HASH_TOP;	/* Toeplitz */
+	rxfh->hfunc = ETH_RSS_HASH_TOP;	/* Toeplitz */
 
 	rndis_dev = ndev->extension;
-	if (indir) {
+	if (rxfh->indir) {
 		for (i = 0; i < ndc->rx_table_sz; i++)
-			indir[i] = ndc->rx_table[i];
+			rxfh->indir[i] = ndc->rx_table[i];
 	}
 
-	if (key)
-		memcpy(key, rndis_dev->rss_key, NETVSC_HASH_KEYLEN);
+	if (rxfh->key)
+		memcpy(rxfh->key, rndis_dev->rss_key, NETVSC_HASH_KEYLEN);
 
 	return 0;
 }
 
-static int netvsc_set_rxfh(struct net_device *dev, const u32 *indir,
-			   const u8 *key, const u8 hfunc)
+static int netvsc_set_rxfh(struct net_device *dev,
+			   struct ethtool_rxfh_param *rxfh,
+			   struct netlink_ext_ack *extack)
 {
 	struct net_device_context *ndc = netdev_priv(dev);
 	struct netvsc_device *ndev = rtnl_dereference(ndc->nvdev);
 	struct rndis_device *rndis_dev;
+	u8 *key = rxfh->key;
 	int i;
 
 	if (!ndev)
 		return -ENODEV;
 
-	if (hfunc != ETH_RSS_HASH_NO_CHANGE && hfunc != ETH_RSS_HASH_TOP)
+	if (rxfh->hfunc != ETH_RSS_HASH_NO_CHANGE &&
+	    rxfh->hfunc != ETH_RSS_HASH_TOP)
 		return -EOPNOTSUPP;
 
 	rndis_dev = ndev->extension;
-	if (indir) {
+	if (rxfh->indir) {
 		for (i = 0; i < ndc->rx_table_sz; i++)
-			if (indir[i] >= ndev->num_chn)
+			if (rxfh->indir[i] >= ndev->num_chn)
 				return -EINVAL;
 
 		for (i = 0; i < ndc->rx_table_sz; i++)
-			ndc->rx_table[i] = indir[i];
+			ndc->rx_table[i] = rxfh->indir[i];
 	}
 
 	if (!key) {
-		if (!indir)
+		if (!rxfh->indir)
 			return 0;
 
 		key = rndis_dev->rss_key;
@@ -2183,7 +2189,7 @@ static rx_handler_result_t netvsc_vf_handle_frame(struct sk_buff **pskb)
 }
 
 static int netvsc_vf_join(struct net_device *vf_netdev,
-			  struct net_device *ndev)
+			  struct net_device *ndev, int context)
 {
 	struct net_device_context *ndev_ctx = netdev_priv(ndev);
 	int ret;
@@ -2206,10 +2212,11 @@ static int netvsc_vf_join(struct net_device *vf_netdev,
 		goto upper_link_failed;
 	}
 
-	/* set slave flag before open to prevent IPv6 addrconf */
-	vf_netdev->flags |= IFF_SLAVE;
-
-	schedule_delayed_work(&ndev_ctx->vf_takeover, VF_TAKEOVER_INT);
+	/* If this registration is called from probe context vf_takeover
+	 * is taken care of later in probe itself.
+	 */
+	if (context == VF_REG_IN_NOTIFIER)
+		schedule_delayed_work(&ndev_ctx->vf_takeover, VF_TAKEOVER_INT);
 
 	call_netdevice_notifiers(NETDEV_JOIN, vf_netdev);
 
@@ -2315,16 +2322,18 @@ static struct net_device *get_netvsc_byslot(const struct net_device *vf_netdev)
 
 	}
 
-	/* Fallback path to check synthetic vf with
-	 * help of mac addr
+	/* Fallback path to check synthetic vf with help of mac addr.
+	 * Because this function can be called before vf_netdev is
+	 * initialized (NETDEV_POST_INIT) when its perm_addr has not been copied
+	 * from dev_addr, also try to match to its dev_addr.
+	 * Note: On Hyper-V and Azure, it's not possible to set a MAC address
+	 * on a VF that matches to the MAC of a unrelated NETVSC device.
 	 */
 	list_for_each_entry(ndev_ctx, &netvsc_dev_list, list) {
 		ndev = hv_get_drvdata(ndev_ctx->device_ctx);
-		if (ether_addr_equal(vf_netdev->perm_addr, ndev->perm_addr)) {
-			netdev_notice(vf_netdev,
-				      "falling back to mac addr based matching\n");
+		if (ether_addr_equal(vf_netdev->perm_addr, ndev->perm_addr) ||
+		    ether_addr_equal(vf_netdev->dev_addr, ndev->perm_addr))
 			return ndev;
-		}
 	}
 
 	netdev_notice(vf_netdev,
@@ -2332,7 +2341,20 @@ static struct net_device *get_netvsc_byslot(const struct net_device *vf_netdev)
 	return NULL;
 }
 
-static int netvsc_register_vf(struct net_device *vf_netdev)
+static int netvsc_prepare_bonding(struct net_device *vf_netdev)
+{
+	struct net_device *ndev;
+
+	ndev = get_netvsc_byslot(vf_netdev);
+	if (!ndev)
+		return NOTIFY_DONE;
+
+	/* set slave flag before open to prevent IPv6 addrconf */
+	vf_netdev->flags |= IFF_SLAVE;
+	return NOTIFY_DONE;
+}
+
+static int netvsc_register_vf(struct net_device *vf_netdev, int context)
 {
 	struct net_device_context *net_device_ctx;
 	struct netvsc_device *netvsc_dev;
@@ -2372,7 +2394,7 @@ static int netvsc_register_vf(struct net_device *vf_netdev)
 
 	netdev_info(ndev, "VF registering: %s\n", vf_netdev->name);
 
-	if (netvsc_vf_join(vf_netdev, ndev) != 0)
+	if (netvsc_vf_join(vf_netdev, ndev, context) != 0)
 		return NOTIFY_DONE;
 
 	dev_hold(vf_netdev);
@@ -2470,10 +2492,31 @@ static int netvsc_unregister_vf(struct net_device *vf_netdev)
 	return NOTIFY_OK;
 }
 
+static int check_dev_is_matching_vf(struct net_device *event_ndev)
+{
+	/* Skip NetVSC interfaces */
+	if (event_ndev->netdev_ops == &device_ops)
+		return -ENODEV;
+
+	/* Avoid non-Ethernet type devices */
+	if (event_ndev->type != ARPHRD_ETHER)
+		return -ENODEV;
+
+	/* Avoid Vlan dev with same MAC registering as VF */
+	if (is_vlan_dev(event_ndev))
+		return -ENODEV;
+
+	/* Avoid Bonding master dev with same MAC registering as VF */
+	if (netif_is_bond_master(event_ndev))
+		return -ENODEV;
+
+	return 0;
+}
+
 static int netvsc_probe(struct hv_device *dev,
 			const struct hv_vmbus_device_id *dev_id)
 {
-	struct net_device *net = NULL;
+	struct net_device *net = NULL, *vf_netdev;
 	struct net_device_context *net_device_ctx;
 	struct netvsc_device_info *device_info = NULL;
 	struct netvsc_device *nvdev;
@@ -2531,15 +2574,6 @@ static int netvsc_probe(struct hv_device *dev,
 		goto devinfo_failed;
 	}
 
-	nvdev = rndis_filter_device_add(dev, device_info);
-	if (IS_ERR(nvdev)) {
-		ret = PTR_ERR(nvdev);
-		netdev_err(net, "unable to add netvsc device (ret %d)\n", ret);
-		goto rndis_failed;
-	}
-
-	eth_hw_addr_set(net, device_info->mac_adr);
-
 	/* We must get rtnl lock before scheduling nvdev->subchan_work,
 	 * otherwise netvsc_subchan_work() can get rtnl lock first and wait
 	 * all subchannels to show up, but that may not happen because
@@ -2547,9 +2581,23 @@ static int netvsc_probe(struct hv_device *dev,
 	 * -> ... -> device_add() -> ... -> __device_attach() can't get
 	 * the device lock, so all the subchannels can't be processed --
 	 * finally netvsc_subchan_work() hangs forever.
+	 *
+	 * The rtnl lock also needs to be held before rndis_filter_device_add()
+	 * which advertises nvsp_2_vsc_capability / sriov bit, and triggers
+	 * VF NIC offering and registering. If VF NIC finished register_netdev()
+	 * earlier it may cause name based config failure.
 	 */
 	rtnl_lock();
 
+	nvdev = rndis_filter_device_add(dev, device_info);
+	if (IS_ERR(nvdev)) {
+		ret = PTR_ERR(nvdev);
+		netdev_err(net, "unable to add netvsc device (ret %d)\n", ret);
+		goto rndis_failed;
+	}
+
+	eth_hw_addr_set(net, device_info->mac_adr);
+
 	if (nvdev->num_chn > 1)
 		schedule_work(&nvdev->subchan_work);
 
@@ -2580,15 +2628,39 @@ static int netvsc_probe(struct hv_device *dev,
 	}
 
 	list_add(&net_device_ctx->list, &netvsc_dev_list);
+
+	/* When the hv_netvsc driver is unloaded and reloaded, the
+	 * NET_DEVICE_REGISTER for the vf device is replayed before probe
+	 * is complete. This is because register_netdevice_notifier() gets
+	 * registered before vmbus_driver_register() so that callback func
+	 * is set before probe and we don't miss events like NETDEV_POST_INIT
+	 * So, in this section we try to register the matching vf device that
+	 * is present as a netdevice, knowing that its register call is not
+	 * processed in the netvsc_netdev_notifier(as probing is progress and
+	 * get_netvsc_byslot fails).
+	 */
+	for_each_netdev(dev_net(net), vf_netdev) {
+		ret = check_dev_is_matching_vf(vf_netdev);
+		if (ret != 0)
+			continue;
+
+		if (net != get_netvsc_byslot(vf_netdev))
+			continue;
+
+		netvsc_prepare_bonding(vf_netdev);
+		netvsc_register_vf(vf_netdev, VF_REG_IN_PROBE);
+		__netvsc_vf_setup(net, vf_netdev);
+		break;
+	}
 	rtnl_unlock();
 
 	netvsc_devinfo_put(device_info);
 	return 0;
 
 register_failed:
-	rtnl_unlock();
 	rndis_filter_device_remove(dev, nvdev);
 rndis_failed:
+	rtnl_unlock();
 	netvsc_devinfo_put(device_info);
 devinfo_failed:
 	free_percpu(net_device_ctx->vf_stats);
@@ -2735,26 +2807,17 @@ static int netvsc_netdev_event(struct notifier_block *this,
 			       unsigned long event, void *ptr)
 {
 	struct net_device *event_dev = netdev_notifier_info_to_dev(ptr);
+	int ret = 0;
 
-	/* Skip our own events */
-	if (event_dev->netdev_ops == &device_ops)
-		return NOTIFY_DONE;
-
-	/* Avoid non-Ethernet type devices */
-	if (event_dev->type != ARPHRD_ETHER)
-		return NOTIFY_DONE;
-
-	/* Avoid Vlan dev with same MAC registering as VF */
-	if (is_vlan_dev(event_dev))
-		return NOTIFY_DONE;
-
-	/* Avoid Bonding master dev with same MAC registering as VF */
-	if (netif_is_bond_master(event_dev))
+	ret = check_dev_is_matching_vf(event_dev);
+	if (ret != 0)
 		return NOTIFY_DONE;
 
 	switch (event) {
+	case NETDEV_POST_INIT:
+		return netvsc_prepare_bonding(event_dev);
 	case NETDEV_REGISTER:
-		return netvsc_register_vf(event_dev);
+		return netvsc_register_vf(event_dev, VF_REG_IN_NOTIFIER);
 	case NETDEV_UNREGISTER:
 		return netvsc_unregister_vf(event_dev);
 	case NETDEV_UP:
@@ -2786,14 +2849,19 @@ static int __init netvsc_drv_init(void)
 		pr_info("Increased ring_size to %u (min allowed)\n",
 			ring_size);
 	}
-	netvsc_ring_bytes = ring_size * PAGE_SIZE;
+	netvsc_ring_bytes = VMBUS_RING_SIZE(ring_size * 4096);
+
+	register_netdevice_notifier(&netvsc_netdev_notifier);
 
 	ret = vmbus_driver_register(&netvsc_drv);
 	if (ret)
-		return ret;
+		goto err_vmbus_reg;
 
-	register_netdevice_notifier(&netvsc_netdev_notifier);
 	return 0;
+
+err_vmbus_reg:
+	unregister_netdevice_notifier(&netvsc_netdev_notifier);
+	return ret;
 }
 
 MODULE_LICENSE("GPL");