From 297b60346df8beafee954a0fd7c2d64f33f3b9bc Mon Sep 17 00:00:00 2001
From: hc <hc@nodka.com>
Date: Sat, 11 May 2024 01:44:05 +0000
Subject: [PATCH] rtl8211F_led_control

---
 kernel/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 3515 ++++++++++++++++++++++++++++++++++++++---------------------
 1 files changed, 2,260 insertions(+), 1,255 deletions(-)

diff --git a/kernel/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/kernel/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 4fb80ed..ae7cd73 100644
--- a/kernel/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/kernel/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -4,31 +4,62 @@
 #include <linux/dma-mapping.h>
 #include <linux/etherdevice.h>
 #include <linux/interrupt.h>
+#ifdef CONFIG_RFS_ACCEL
+#include <linux/cpu_rmap.h>
+#endif
 #include <linux/if_vlan.h>
+#include <linux/irq.h>
 #include <linux/ip.h>
 #include <linux/ipv6.h>
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/aer.h>
 #include <linux/skbuff.h>
 #include <linux/sctp.h>
-#include <linux/vermagic.h>
 #include <net/gre.h>
+#include <net/ip6_checksum.h>
 #include <net/pkt_cls.h>
+#include <net/tcp.h>
 #include <net/vxlan.h>
+#include <net/geneve.h>
 
 #include "hnae3.h"
 #include "hns3_enet.h"
+/* All hns3 tracepoints are defined by the include below, which
+ * must be included exactly once across the whole kernel with
+ * CREATE_TRACE_POINTS defined
+ */
+#define CREATE_TRACE_POINTS
+#include "hns3_trace.h"
 
-static void hns3_clear_all_ring(struct hnae3_handle *h);
-static void hns3_force_clear_all_rx_ring(struct hnae3_handle *h);
+#define hns3_set_field(origin, shift, val)	((origin) |= ((val) << (shift)))
+#define hns3_tx_bd_count(S)	DIV_ROUND_UP(S, HNS3_MAX_BD_SIZE)
+
+#define hns3_rl_err(fmt, ...)						\
+	do {								\
+		if (net_ratelimit())					\
+			netdev_err(fmt, ##__VA_ARGS__);			\
+	} while (0)
+
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force);
 
 static const char hns3_driver_name[] = "hns3";
-const char hns3_driver_version[] = VERMAGIC_STRING;
 static const char hns3_driver_string[] =
 			"Hisilicon Ethernet Network Driver for Hip08 Family";
 static const char hns3_copyright[] = "Copyright (c) 2017 Huawei Corporation.";
 static struct hnae3_client client;
 
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, " Network interface message level setting");
+
+#define DEFAULT_MSG_LEVEL (NETIF_MSG_PROBE | NETIF_MSG_LINK | \
+			   NETIF_MSG_IFDOWN | NETIF_MSG_IFUP)
+
+#define HNS3_INNER_VLAN_TAG	1
+#define HNS3_OUTER_VLAN_TAG	2
+
+#define HNS3_MIN_TX_LEN		33U
 #define HNS3_MIN_TUN_PKT_LEN	65U
 
 /* hns3_pci_tbl - PCI Device ID Table
@@ -51,8 +82,10 @@
 	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
 	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_MACSEC),
 	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
-	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_VF), 0},
-	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF),
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_200G_RDMA),
+	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_VF), 0},
+	{PCI_VDEVICE(HUAWEI, HNAE3_DEV_ID_RDMA_DCB_PFC_VF),
 	 HNAE3_DEV_SUPPORT_ROCE_DCB_BITS},
 	/* required last entry */
 	{0, }
@@ -63,7 +96,7 @@
 {
 	struct hns3_enet_tqp_vector *tqp_vector = vector;
 
-	napi_schedule(&tqp_vector->napi);
+	napi_schedule_irqoff(&tqp_vector->napi);
 
 	return IRQ_HANDLED;
 }
@@ -78,6 +111,9 @@
 
 		if (tqp_vectors->irq_init_flag != HNS3_VECTOR_INITED)
 			continue;
+
+		/* clear the affinity mask */
+		irq_set_affinity_hint(tqp_vectors->vector_irq, NULL);
 
 		/* release the irq resource */
 		free_irq(tqp_vectors->vector_irq, tqp_vectors);
@@ -101,18 +137,21 @@
 			continue;
 
 		if (tqp_vectors->tx_group.ring && tqp_vectors->rx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "TxRx",
-				 txrx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "TxRx", txrx_int_idx++);
 			txrx_int_idx++;
 		} else if (tqp_vectors->rx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "Rx",
-				 rx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Rx", rx_int_idx++);
 		} else if (tqp_vectors->tx_group.ring) {
-			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN - 1,
-				 "%s-%s-%d", priv->netdev->name, "Tx",
-				 tx_int_idx++);
+			snprintf(tqp_vectors->name, HNAE3_INT_NAME_LEN,
+				 "%s-%s-%s-%d", hns3_driver_name,
+				 pci_name(priv->ae_handle->pdev),
+				 "Tx", tx_int_idx++);
 		} else {
 			/* Skip this unused q_vector */
 			continue;
@@ -120,14 +159,18 @@
 
 		tqp_vectors->name[HNAE3_INT_NAME_LEN - 1] = '\0';
 
+		irq_set_status_flags(tqp_vectors->vector_irq, IRQ_NOAUTOEN);
 		ret = request_irq(tqp_vectors->vector_irq, hns3_irq_handle, 0,
-				  tqp_vectors->name,
-				       tqp_vectors);
+				  tqp_vectors->name, tqp_vectors);
 		if (ret) {
 			netdev_err(priv->netdev, "request irq(%d) fail\n",
 				   tqp_vectors->vector_irq);
+			hns3_nic_uninit_irq(priv);
 			return ret;
 		}
+
+		irq_set_affinity_hint(tqp_vectors->vector_irq,
+				      &tqp_vectors->affinity_mask);
 
 		tqp_vectors->irq_init_flag = HNS3_VECTOR_INITED;
 	}
@@ -144,6 +187,7 @@
 static void hns3_vector_enable(struct hns3_enet_tqp_vector *tqp_vector)
 {
 	napi_enable(&tqp_vector->napi);
+	enable_irq(tqp_vector->vector_irq);
 
 	/* enable vector */
 	hns3_mask_vector_irq(tqp_vector, 1);
@@ -200,16 +244,15 @@
 	/* initialize the configuration for interrupt coalescing.
 	 * 1. GL (Interrupt Gap Limiter)
 	 * 2. RL (Interrupt Rate Limiter)
+	 *
+	 * Default: enable interrupt coalescing self-adaptive and GL
 	 */
-
-	/* Default: enable interrupt coalescing self-adaptive and GL */
 	tqp_vector->tx_group.coal.gl_adapt_enable = 1;
 	tqp_vector->rx_group.coal.gl_adapt_enable = 1;
 
 	tqp_vector->tx_group.coal.int_gl = HNS3_INT_GL_50K;
 	tqp_vector->rx_group.coal.int_gl = HNS3_INT_GL_50K;
 
-	tqp_vector->int_adapt_down = HNS3_INT_ADAPT_DOWN_START;
 	tqp_vector->rx_group.coal.flow_level = HNS3_FLOW_LOW;
 	tqp_vector->tx_group.coal.flow_level = HNS3_FLOW_LOW;
 }
@@ -257,8 +300,7 @@
 	ret = netif_set_real_num_tx_queues(netdev, queue_size);
 	if (ret) {
 		netdev_err(netdev,
-			   "netif_set_real_num_tx_queues fail, ret=%d!\n",
-			   ret);
+			   "netif_set_real_num_tx_queues fail, ret=%d!\n", ret);
 		return ret;
 	}
 
@@ -274,12 +316,64 @@
 
 static u16 hns3_get_max_available_channels(struct hnae3_handle *h)
 {
-	u16 free_tqps, max_rss_size, max_tqps;
+	u16 alloc_tqps, max_rss_size, rss_size;
 
-	h->ae_algo->ops->get_tqps_and_rss_info(h, &free_tqps, &max_rss_size);
-	max_tqps = h->kinfo.num_tc * max_rss_size;
+	h->ae_algo->ops->get_tqps_and_rss_info(h, &alloc_tqps, &max_rss_size);
+	rss_size = alloc_tqps / h->kinfo.num_tc;
 
-	return min_t(u16, max_tqps, (free_tqps + h->kinfo.num_tqps));
+	return min_t(u16, rss_size, max_rss_size);
+}
+
+static void hns3_tqp_enable(struct hnae3_queue *tqp)
+{
+	u32 rcb_reg;
+
+	rcb_reg = hns3_read_dev(tqp, HNS3_RING_EN_REG);
+	rcb_reg |= BIT(HNS3_RING_EN_B);
+	hns3_write_dev(tqp, HNS3_RING_EN_REG, rcb_reg);
+}
+
+static void hns3_tqp_disable(struct hnae3_queue *tqp)
+{
+	u32 rcb_reg;
+
+	rcb_reg = hns3_read_dev(tqp, HNS3_RING_EN_REG);
+	rcb_reg &= ~BIT(HNS3_RING_EN_B);
+	hns3_write_dev(tqp, HNS3_RING_EN_REG, rcb_reg);
+}
+
+static void hns3_free_rx_cpu_rmap(struct net_device *netdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	free_irq_cpu_rmap(netdev->rx_cpu_rmap);
+	netdev->rx_cpu_rmap = NULL;
+#endif
+}
+
+static int hns3_set_rx_cpu_rmap(struct net_device *netdev)
+{
+#ifdef CONFIG_RFS_ACCEL
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int i, ret;
+
+	if (!netdev->rx_cpu_rmap) {
+		netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(priv->vector_num);
+		if (!netdev->rx_cpu_rmap)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < priv->vector_num; i++) {
+		tqp_vector = &priv->tqp_vector[i];
+		ret = irq_cpu_rmap_add(netdev->rx_cpu_rmap,
+				       tqp_vector->vector_irq);
+		if (ret) {
+			hns3_free_rx_cpu_rmap(netdev);
+			return ret;
+		}
+	}
+#endif
+	return 0;
 }
 
 static int hns3_nic_net_up(struct net_device *netdev)
@@ -293,33 +387,51 @@
 	if (ret)
 		return ret;
 
-	/* get irq resource for all vectors */
-	ret = hns3_nic_init_irq(priv);
-	if (ret) {
-		netdev_err(netdev, "hns init irq failed! ret=%d\n", ret);
-		return ret;
-	}
+	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
 
 	/* enable the vectors */
 	for (i = 0; i < priv->vector_num; i++)
 		hns3_vector_enable(&priv->tqp_vector[i]);
 
+	/* enable rcb */
+	for (j = 0; j < h->kinfo.num_tqps; j++)
+		hns3_tqp_enable(h->kinfo.tqp[j]);
+
 	/* start the ae_dev */
 	ret = h->ae_algo->ops->start ? h->ae_algo->ops->start(h) : 0;
-	if (ret)
-		goto out_start_err;
+	if (ret) {
+		set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+		while (j--)
+			hns3_tqp_disable(h->kinfo.tqp[j]);
 
-	clear_bit(HNS3_NIC_STATE_DOWN, &priv->state);
-
-	return 0;
-
-out_start_err:
-	for (j = i - 1; j >= 0; j--)
-		hns3_vector_disable(&priv->tqp_vector[j]);
-
-	hns3_nic_uninit_irq(priv);
+		for (j = i - 1; j >= 0; j--)
+			hns3_vector_disable(&priv->tqp_vector[j]);
+	}
 
 	return ret;
+}
+
+static void hns3_config_xps(struct hns3_nic_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < priv->vector_num; i++) {
+		struct hns3_enet_tqp_vector *tqp_vector = &priv->tqp_vector[i];
+		struct hns3_enet_ring *ring = tqp_vector->tx_group.ring;
+
+		while (ring) {
+			int ret;
+
+			ret = netif_set_xps_queue(priv->netdev,
+						  &tqp_vector->affinity_mask,
+						  ring->tqp->tqp_index);
+			if (ret)
+				netdev_warn(priv->netdev,
+					    "set xps queue failed: %d", ret);
+
+			ring = ring->next;
+		}
+	}
 }
 
 static int hns3_nic_net_open(struct net_device *netdev)
@@ -329,6 +441,14 @@
 	struct hnae3_knic_private_info *kinfo;
 	int i, ret;
 
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (!test_bit(HNS3_NIC_STATE_DOWN, &priv->state)) {
+		netdev_warn(netdev, "net open repeatedly!\n");
+		return 0;
+	}
+
 	netif_carrier_off(netdev);
 
 	ret = hns3_nic_set_real_num_queue(netdev);
@@ -337,49 +457,83 @@
 
 	ret = hns3_nic_net_up(netdev);
 	if (ret) {
-		netdev_err(netdev,
-			   "hns net up fail, ret=%d!\n", ret);
+		netdev_err(netdev, "net up fail, ret=%d!\n", ret);
 		return ret;
 	}
 
 	kinfo = &h->kinfo;
-	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++) {
-		netdev_set_prio_tc_map(netdev, i,
-				       kinfo->prio_tc[i]);
-	}
+	for (i = 0; i < HNAE3_MAX_USER_PRIO; i++)
+		netdev_set_prio_tc_map(netdev, i, kinfo->prio_tc[i]);
 
-	priv->ae_handle->last_reset_time = jiffies;
+	if (h->ae_algo->ops->set_timer_task)
+		h->ae_algo->ops->set_timer_task(priv->ae_handle, true);
+
+	hns3_config_xps(priv);
+
+	netif_dbg(h, drv, netdev, "net open\n");
+
 	return 0;
+}
+
+static void hns3_reset_tx_queue(struct hnae3_handle *h)
+{
+	struct net_device *ndev = h->kinfo.netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+	struct netdev_queue *dev_queue;
+	u32 i;
+
+	for (i = 0; i < h->kinfo.num_tqps; i++) {
+		dev_queue = netdev_get_tx_queue(ndev,
+						priv->ring[i].queue_index);
+		netdev_tx_reset_queue(dev_queue);
+	}
 }
 
 static void hns3_nic_net_down(struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
 	const struct hnae3_ae_ops *ops;
 	int i;
-
-	if (test_and_set_bit(HNS3_NIC_STATE_DOWN, &priv->state))
-		return;
 
 	/* disable vectors */
 	for (i = 0; i < priv->vector_num; i++)
 		hns3_vector_disable(&priv->tqp_vector[i]);
+
+	/* disable rcb */
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_tqp_disable(h->kinfo.tqp[i]);
 
 	/* stop ae_dev */
 	ops = priv->ae_handle->ae_algo->ops;
 	if (ops->stop)
 		ops->stop(priv->ae_handle);
 
-	/* free irq resources */
-	hns3_nic_uninit_irq(priv);
+	/* delay ring buffer clearing to hns3_reset_notify_uninit_enet
+	 * during reset process, because driver may not be able
+	 * to disable the ring through firmware when downing the netdev.
+	 */
+	if (!hns3_nic_resetting(netdev))
+		hns3_clear_all_ring(priv->ae_handle, false);
 
-	hns3_clear_all_ring(priv->ae_handle);
+	hns3_reset_tx_queue(priv->ae_handle);
 }
 
 static int hns3_nic_net_stop(struct net_device *netdev)
 {
-	netif_tx_stop_all_queues(netdev);
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (test_and_set_bit(HNS3_NIC_STATE_DOWN, &priv->state))
+		return 0;
+
+	netif_dbg(h, drv, netdev, "net stop\n");
+
+	if (h->ae_algo->ops->set_timer_task)
+		h->ae_algo->ops->set_timer_task(priv->ae_handle, false);
+
 	netif_carrier_off(netdev);
+	netif_tx_disable(netdev);
 
 	hns3_nic_net_down(netdev);
 
@@ -401,6 +555,13 @@
 			      const unsigned char *addr)
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	/* need ignore the request of removing device address, because
+	 * we store the device address and other addresses of uc list
+	 * in the function's mac filter list.
+	 */
+	if (ether_addr_equal(addr, netdev->dev_addr))
+		return 0;
 
 	if (h->ae_algo->ops->rm_uc_addr)
 		return h->ae_algo->ops->rm_uc_addr(h, addr);
@@ -430,26 +591,62 @@
 	return 0;
 }
 
+static u8 hns3_get_netdev_flags(struct net_device *netdev)
+{
+	u8 flags = 0;
+
+	if (netdev->flags & IFF_PROMISC) {
+		flags = HNAE3_USER_UPE | HNAE3_USER_MPE | HNAE3_BPE;
+	} else {
+		flags |= HNAE3_VLAN_FLTR;
+		if (netdev->flags & IFF_ALLMULTI)
+			flags |= HNAE3_USER_MPE;
+	}
+
+	return flags;
+}
+
 static void hns3_nic_set_rx_mode(struct net_device *netdev)
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
+	u8 new_flags;
 
-	if (h->ae_algo->ops->set_promisc_mode) {
-		if (netdev->flags & IFF_PROMISC)
-			h->ae_algo->ops->set_promisc_mode(h, true, true);
-		else if (netdev->flags & IFF_ALLMULTI)
-			h->ae_algo->ops->set_promisc_mode(h, false, true);
-		else
-			h->ae_algo->ops->set_promisc_mode(h, false, false);
-	}
-	if (__dev_uc_sync(netdev, hns3_nic_uc_sync, hns3_nic_uc_unsync))
-		netdev_err(netdev, "sync uc address fail\n");
-	if (netdev->flags & IFF_MULTICAST) {
-		if (__dev_mc_sync(netdev, hns3_nic_mc_sync, hns3_nic_mc_unsync))
-			netdev_err(netdev, "sync mc address fail\n");
+	new_flags = hns3_get_netdev_flags(netdev);
 
-		if (h->ae_algo->ops->update_mta_status)
-			h->ae_algo->ops->update_mta_status(h);
+	__dev_uc_sync(netdev, hns3_nic_uc_sync, hns3_nic_uc_unsync);
+	__dev_mc_sync(netdev, hns3_nic_mc_sync, hns3_nic_mc_unsync);
+
+	/* User mode Promisc mode enable and vlan filtering is disabled to
+	 * let all packets in.
+	 */
+	h->netdev_flags = new_flags;
+	hns3_request_update_promisc_mode(h);
+}
+
+void hns3_request_update_promisc_mode(struct hnae3_handle *handle)
+{
+	const struct hnae3_ae_ops *ops = handle->ae_algo->ops;
+
+	if (ops->request_update_promisc_mode)
+		ops->request_update_promisc_mode(handle);
+}
+
+void hns3_enable_vlan_filter(struct net_device *netdev, bool enable)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = priv->ae_handle;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(h->pdev);
+	bool last_state;
+
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2 &&
+	    h->ae_algo->ops->enable_vlan_filter) {
+		last_state = h->netdev_flags & HNAE3_VLAN_FLTR ? true : false;
+		if (enable != last_state) {
+			netdev_info(netdev,
+				    "%s vlan filter\n",
+				    enable ? "enable" : "disable");
+			h->ae_algo->ops->enable_vlan_filter(h, enable);
+		}
 	}
 }
 
@@ -466,7 +663,7 @@
 		return 0;
 
 	ret = skb_cow_head(skb, 0);
-	if (ret)
+	if (unlikely(ret < 0))
 		return ret;
 
 	l3.hdr = skb_network_header(skb);
@@ -478,7 +675,7 @@
 	if (l3.v4->version == 4)
 		l3.v4->check = 0;
 
-	/* tunnel packet.*/
+	/* tunnel packet */
 	if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE |
 					 SKB_GSO_GRE_CSUM |
 					 SKB_GSO_UDP_TUNNEL |
@@ -503,22 +700,30 @@
 			l3.v4->check = 0;
 	}
 
-	/* normal or tunnel packet*/
+	/* normal or tunnel packet */
 	l4_offset = l4.hdr - skb->data;
-	hdr_len = (l4.tcp->doff * 4) + l4_offset;
 
-	/* remove payload length from inner pseudo checksum when tso*/
+	/* remove payload length from inner pseudo checksum when tso */
 	l4_paylen = skb->len - l4_offset;
-	csum_replace_by_diff(&l4.tcp->check,
-			     (__force __wsum)htonl(l4_paylen));
+
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
+		hdr_len = sizeof(*l4.udp) + l4_offset;
+		csum_replace_by_diff(&l4.udp->check,
+				     (__force __wsum)htonl(l4_paylen));
+	} else {
+		hdr_len = (l4.tcp->doff << 2) + l4_offset;
+		csum_replace_by_diff(&l4.tcp->check,
+				     (__force __wsum)htonl(l4_paylen));
+	}
 
 	/* find the txbd field values */
 	*paylen = skb->len - hdr_len;
-	hnae3_set_bit(*type_cs_vlan_tso,
-		      HNS3_TXD_TSO_B, 1);
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_TSO_B, 1);
 
 	/* get MSS for TSO */
 	*mss = skb_shinfo(skb)->gso_size;
+
+	trace_hns3_tso(skb);
 
 	return 0;
 }
@@ -526,11 +731,7 @@
 static int hns3_get_l4_protocol(struct sk_buff *skb, u8 *ol4_proto,
 				u8 *il4_proto)
 {
-	union {
-		struct iphdr *v4;
-		struct ipv6hdr *v6;
-		unsigned char *hdr;
-	} l3;
+	union l3_hdr_info l3;
 	unsigned char *l4_hdr;
 	unsigned char *exthdr;
 	u8 l4_proto_tmp;
@@ -579,182 +780,92 @@
 	return 0;
 }
 
-static void hns3_set_l2l3l4_len(struct sk_buff *skb, u8 ol4_proto,
-				u8 il4_proto, u32 *type_cs_vlan_tso,
-				u32 *ol_type_vlan_len_msec)
-{
-	union {
-		struct iphdr *v4;
-		struct ipv6hdr *v6;
-		unsigned char *hdr;
-	} l3;
-	union {
-		struct tcphdr *tcp;
-		struct udphdr *udp;
-		struct gre_base_hdr *gre;
-		unsigned char *hdr;
-	} l4;
-	unsigned char *l2_hdr;
-	u8 l4_proto = ol4_proto;
-	u32 ol2_len;
-	u32 ol3_len;
-	u32 ol4_len;
-	u32 l2_len;
-	u32 l3_len;
-
-	l3.hdr = skb_network_header(skb);
-	l4.hdr = skb_transport_header(skb);
-
-	/* compute L2 header size for normal packet, defined in 2 Bytes */
-	l2_len = l3.hdr - skb->data;
-	hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L2LEN_M,
-			HNS3_TXD_L2LEN_S, l2_len >> 1);
-
-	/* tunnel packet*/
-	if (skb->encapsulation) {
-		/* compute OL2 header size, defined in 2 Bytes */
-		ol2_len = l2_len;
-		hnae3_set_field(*ol_type_vlan_len_msec,
-				HNS3_TXD_L2LEN_M,
-				HNS3_TXD_L2LEN_S, ol2_len >> 1);
-
-		/* compute OL3 header size, defined in 4 Bytes */
-		ol3_len = l4.hdr - l3.hdr;
-		hnae3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L3LEN_M,
-				HNS3_TXD_L3LEN_S, ol3_len >> 2);
-
-		/* MAC in UDP, MAC in GRE (0x6558)*/
-		if ((ol4_proto == IPPROTO_UDP) || (ol4_proto == IPPROTO_GRE)) {
-			/* switch MAC header ptr from outer to inner header.*/
-			l2_hdr = skb_inner_mac_header(skb);
-
-			/* compute OL4 header size, defined in 4 Bytes. */
-			ol4_len = l2_hdr - l4.hdr;
-			hnae3_set_field(*ol_type_vlan_len_msec,
-					HNS3_TXD_L4LEN_M, HNS3_TXD_L4LEN_S,
-					ol4_len >> 2);
-
-			/* switch IP header ptr from outer to inner header */
-			l3.hdr = skb_inner_network_header(skb);
-
-			/* compute inner l2 header size, defined in 2 Bytes. */
-			l2_len = l3.hdr - l2_hdr;
-			hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L2LEN_M,
-					HNS3_TXD_L2LEN_S, l2_len >> 1);
-		} else {
-			/* skb packet types not supported by hardware,
-			 * txbd len fild doesn't be filled.
-			 */
-			return;
-		}
-
-		/* switch L4 header pointer from outer to inner */
-		l4.hdr = skb_inner_transport_header(skb);
-
-		l4_proto = il4_proto;
-	}
-
-	/* compute inner(/normal) L3 header size, defined in 4 Bytes */
-	l3_len = l4.hdr - l3.hdr;
-	hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3LEN_M,
-			HNS3_TXD_L3LEN_S, l3_len >> 2);
-
-	/* compute inner(/normal) L4 header size, defined in 4 Bytes */
-	switch (l4_proto) {
-	case IPPROTO_TCP:
-		hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_M,
-				HNS3_TXD_L4LEN_S, l4.tcp->doff);
-		break;
-	case IPPROTO_SCTP:
-		hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_M,
-				HNS3_TXD_L4LEN_S,
-				(sizeof(struct sctphdr) >> 2));
-		break;
-	case IPPROTO_UDP:
-		hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_M,
-				HNS3_TXD_L4LEN_S,
-				(sizeof(struct udphdr) >> 2));
-		break;
-	default:
-		/* skb packet types not supported by hardware,
-		 * txbd len fild doesn't be filled.
-		 */
-		return;
-	}
-}
-
 /* when skb->encapsulation is 0, skb->ip_summed is CHECKSUM_PARTIAL
  * and it is udp packet, which has a dest port as the IANA assigned.
  * the hardware is expected to do the checksum offload, but the
  * hardware will not do the checksum offload when udp dest port is
- * 4789.
+ * 4789, 4790 or 6081.
  */
 static bool hns3_tunnel_csum_bug(struct sk_buff *skb)
 {
-#define IANA_VXLAN_PORT	4789
-	union {
-		struct tcphdr *tcp;
-		struct udphdr *udp;
-		struct gre_base_hdr *gre;
-		unsigned char *hdr;
-	} l4;
+	union l4_hdr_info l4;
 
 	l4.hdr = skb_transport_header(skb);
 
-	if (!(!skb->encapsulation && l4.udp->dest == htons(IANA_VXLAN_PORT)))
+	if (!(!skb->encapsulation &&
+	      (l4.udp->dest == htons(IANA_VXLAN_UDP_PORT) ||
+	      l4.udp->dest == htons(GENEVE_UDP_PORT) ||
+	      l4.udp->dest == htons(4790))))
 		return false;
 
 	return true;
 }
 
-static int hns3_set_l3l4_type_csum(struct sk_buff *skb, u8 ol4_proto,
-				   u8 il4_proto, u32 *type_cs_vlan_tso,
-				   u32 *ol_type_vlan_len_msec)
+static void hns3_set_outer_l2l3l4(struct sk_buff *skb, u8 ol4_proto,
+				  u32 *ol_type_vlan_len_msec)
 {
-	union {
-		struct iphdr *v4;
-		struct ipv6hdr *v6;
-		unsigned char *hdr;
-	} l3;
-	u32 l4_proto = ol4_proto;
+	u32 l2_len, l3_len, l4_len;
+	unsigned char *il2_hdr;
+	union l3_hdr_info l3;
+	union l4_hdr_info l4;
 
 	l3.hdr = skb_network_header(skb);
+	l4.hdr = skb_transport_header(skb);
 
-	/* define OL3 type and tunnel type(OL4).*/
+	/* compute OL2 header size, defined in 2 Bytes */
+	l2_len = l3.hdr - skb->data;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L2LEN_S, l2_len >> 1);
+
+	/* compute OL3 header size, defined in 4 Bytes */
+	l3_len = l4.hdr - l3.hdr;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L3LEN_S, l3_len >> 2);
+
+	il2_hdr = skb_inner_mac_header(skb);
+	/* compute OL4 header size, defined in 4 Bytes */
+	l4_len = il2_hdr - l4.hdr;
+	hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_L4LEN_S, l4_len >> 2);
+
+	/* define outer network header type */
+	if (skb->protocol == htons(ETH_P_IP)) {
+		if (skb_is_gso(skb))
+			hns3_set_field(*ol_type_vlan_len_msec,
+				       HNS3_TXD_OL3T_S,
+				       HNS3_OL3T_IPV4_CSUM);
+		else
+			hns3_set_field(*ol_type_vlan_len_msec,
+				       HNS3_TXD_OL3T_S,
+				       HNS3_OL3T_IPV4_NO_CSUM);
+
+	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_OL3T_S,
+			       HNS3_OL3T_IPV6);
+	}
+
+	if (ol4_proto == IPPROTO_UDP)
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_TUNTYPE_S,
+			       HNS3_TUN_MAC_IN_UDP);
+	else if (ol4_proto == IPPROTO_GRE)
+		hns3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_TUNTYPE_S,
+			       HNS3_TUN_NVGRE);
+}
+
+static int hns3_set_l2l3l4(struct sk_buff *skb, u8 ol4_proto,
+			   u8 il4_proto, u32 *type_cs_vlan_tso,
+			   u32 *ol_type_vlan_len_msec)
+{
+	unsigned char *l2_hdr = skb->data;
+	u32 l4_proto = ol4_proto;
+	union l4_hdr_info l4;
+	union l3_hdr_info l3;
+	u32 l2_len, l3_len;
+
+	l4.hdr = skb_transport_header(skb);
+	l3.hdr = skb_network_header(skb);
+
+	/* handle encapsulation skb */
 	if (skb->encapsulation) {
-		/* define outer network header type.*/
-		if (skb->protocol == htons(ETH_P_IP)) {
-			if (skb_is_gso(skb))
-				hnae3_set_field(*ol_type_vlan_len_msec,
-						HNS3_TXD_OL3T_M,
-						HNS3_TXD_OL3T_S,
-						HNS3_OL3T_IPV4_CSUM);
-			else
-				hnae3_set_field(*ol_type_vlan_len_msec,
-						HNS3_TXD_OL3T_M,
-						HNS3_TXD_OL3T_S,
-						HNS3_OL3T_IPV4_NO_CSUM);
-
-		} else if (skb->protocol == htons(ETH_P_IPV6)) {
-			hnae3_set_field(*ol_type_vlan_len_msec, HNS3_TXD_OL3T_M,
-					HNS3_TXD_OL3T_S, HNS3_OL3T_IPV6);
-		}
-
-		/* define tunnel type(OL4).*/
-		switch (l4_proto) {
-		case IPPROTO_UDP:
-			hnae3_set_field(*ol_type_vlan_len_msec,
-					HNS3_TXD_TUNTYPE_M,
-					HNS3_TXD_TUNTYPE_S,
-					HNS3_TUN_MAC_IN_UDP);
-			break;
-		case IPPROTO_GRE:
-			hnae3_set_field(*ol_type_vlan_len_msec,
-					HNS3_TXD_TUNTYPE_M,
-					HNS3_TXD_TUNTYPE_S,
-					HNS3_TUN_NVGRE);
-			break;
-		default:
+		/* If this is a not UDP/GRE encapsulation skb */
+		if (!(ol4_proto == IPPROTO_UDP || ol4_proto == IPPROTO_GRE)) {
 			/* drop the skb tunnel packet if hardware don't support,
 			 * because hardware can't calculate csum when TSO.
 			 */
@@ -767,31 +878,45 @@
 			return skb_checksum_help(skb);
 		}
 
+		hns3_set_outer_l2l3l4(skb, ol4_proto, ol_type_vlan_len_msec);
+
+		/* switch to inner header */
+		l2_hdr = skb_inner_mac_header(skb);
 		l3.hdr = skb_inner_network_header(skb);
+		l4.hdr = skb_inner_transport_header(skb);
 		l4_proto = il4_proto;
 	}
 
 	if (l3.v4->version == 4) {
-		hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_M,
-				HNS3_TXD_L3T_S, HNS3_L3T_IPV4);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_S,
+			       HNS3_L3T_IPV4);
 
 		/* the stack computes the IP header already, the only time we
 		 * need the hardware to recompute it is in the case of TSO.
 		 */
 		if (skb_is_gso(skb))
-			hnae3_set_bit(*type_cs_vlan_tso, HNS3_TXD_L3CS_B, 1);
+			hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3CS_B, 1);
 	} else if (l3.v6->version == 6) {
-		hnae3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_M,
-				HNS3_TXD_L3T_S, HNS3_L3T_IPV6);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3T_S,
+			       HNS3_L3T_IPV6);
 	}
 
+	/* compute inner(/normal) L2 header size, defined in 2 Bytes */
+	l2_len = l3.hdr - l2_hdr;
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L2LEN_S, l2_len >> 1);
+
+	/* compute inner(/normal) L3 header size, defined in 4 Bytes */
+	l3_len = l4.hdr - l3.hdr;
+	hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L3LEN_S, l3_len >> 2);
+
+	/* compute inner(/normal) L4 header size, defined in 4 Bytes */
 	switch (l4_proto) {
 	case IPPROTO_TCP:
-		hnae3_set_bit(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
-		hnae3_set_field(*type_cs_vlan_tso,
-				HNS3_TXD_L4T_M,
-				HNS3_TXD_L4T_S,
-				HNS3_L4T_TCP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_TCP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       l4.tcp->doff);
 		break;
 	case IPPROTO_UDP:
 		if (hns3_tunnel_csum_bug(skb)) {
@@ -800,18 +925,18 @@
 			return ret ? ret : skb_checksum_help(skb);
 		}
 
-		hnae3_set_bit(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
-		hnae3_set_field(*type_cs_vlan_tso,
-				HNS3_TXD_L4T_M,
-				HNS3_TXD_L4T_S,
-				HNS3_L4T_UDP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_UDP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       (sizeof(struct udphdr) >> 2));
 		break;
 	case IPPROTO_SCTP:
-		hnae3_set_bit(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
-		hnae3_set_field(*type_cs_vlan_tso,
-				HNS3_TXD_L4T_M,
-				HNS3_TXD_L4T_S,
-				HNS3_L4T_SCTP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4CS_B, 1);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4T_S,
+			       HNS3_L4T_SCTP);
+		hns3_set_field(*type_cs_vlan_tso, HNS3_TXD_L4LEN_S,
+			       (sizeof(struct sctphdr) >> 2));
 		break;
 	default:
 		/* drop the skb tunnel packet if hardware don't support,
@@ -829,28 +954,27 @@
 	return 0;
 }
 
-static void hns3_set_txbd_baseinfo(u16 *bdtp_fe_sc_vld_ra_ri, int frag_end)
+static int hns3_handle_vtags(struct hns3_enet_ring *tx_ring,
+			     struct sk_buff *skb)
 {
-	/* Config bd buffer end */
-	hnae3_set_field(*bdtp_fe_sc_vld_ra_ri, HNS3_TXD_BDTYPE_M,
-			HNS3_TXD_BDTYPE_S, 0);
-	hnae3_set_bit(*bdtp_fe_sc_vld_ra_ri, HNS3_TXD_FE_B, !!frag_end);
-	hnae3_set_bit(*bdtp_fe_sc_vld_ra_ri, HNS3_TXD_VLD_B, 1);
-	hnae3_set_field(*bdtp_fe_sc_vld_ra_ri, HNS3_TXD_SC_M, HNS3_TXD_SC_S, 0);
-}
+	struct hnae3_handle *handle = tx_ring->tqp->handle;
+	struct vlan_ethhdr *vhdr;
+	int rc;
 
-static int hns3_fill_desc_vtags(struct sk_buff *skb,
-				struct hns3_enet_ring *tx_ring,
-				u32 *inner_vlan_flag,
-				u32 *out_vlan_flag,
-				u16 *inner_vtag,
-				u16 *out_vtag)
-{
-#define HNS3_TX_VLAN_PRIO_SHIFT 13
+	if (!(skb->protocol == htons(ETH_P_8021Q) ||
+	      skb_vlan_tag_present(skb)))
+		return 0;
+
+	/* Since HW limitation, if port based insert VLAN enabled, only one VLAN
+	 * header is allowed in skb, otherwise it will cause RAS error.
+	 */
+	if (unlikely(skb_vlan_tagged_multi(skb) &&
+		     handle->port_base_vlan_state ==
+		     HNAE3_PORT_BASE_VLAN_ENABLE))
+		return -EINVAL;
 
 	if (skb->protocol == htons(ETH_P_8021Q) &&
-	    !(tx_ring->tqp->handle->kinfo.netdev->features &
-	    NETIF_F_HW_VLAN_CTAG_TX)) {
+	    !(handle->kinfo.netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) {
 		/* When HW VLAN acceleration is turned off, and the stack
 		 * sets the protocol to 802.1q, the driver just need to
 		 * set the protocol to the encapsulated ethertype.
@@ -860,323 +984,556 @@
 	}
 
 	if (skb_vlan_tag_present(skb)) {
-		u16 vlan_tag;
-
-		vlan_tag = skb_vlan_tag_get(skb);
-		vlan_tag |= (skb->priority & 0x7) << HNS3_TX_VLAN_PRIO_SHIFT;
-
 		/* Based on hw strategy, use out_vtag in two layer tag case,
 		 * and use inner_vtag in one tag case.
 		 */
-		if (skb->protocol == htons(ETH_P_8021Q)) {
-			hnae3_set_bit(*out_vlan_flag, HNS3_TXD_OVLAN_B, 1);
-			*out_vtag = vlan_tag;
-		} else {
-			hnae3_set_bit(*inner_vlan_flag, HNS3_TXD_VLAN_B, 1);
-			*inner_vtag = vlan_tag;
-		}
-	} else if (skb->protocol == htons(ETH_P_8021Q)) {
-		struct vlan_ethhdr *vhdr;
-		int rc;
+		if (skb->protocol == htons(ETH_P_8021Q) &&
+		    handle->port_base_vlan_state ==
+		    HNAE3_PORT_BASE_VLAN_DISABLE)
+			rc = HNS3_OUTER_VLAN_TAG;
+		else
+			rc = HNS3_INNER_VLAN_TAG;
 
-		rc = skb_cow_head(skb, 0);
-		if (rc < 0)
-			return rc;
-		vhdr = (struct vlan_ethhdr *)skb->data;
-		vhdr->h_vlan_TCI |= cpu_to_be16((skb->priority & 0x7)
-					<< HNS3_TX_VLAN_PRIO_SHIFT);
+		skb->protocol = vlan_get_protocol(skb);
+		return rc;
 	}
+
+	rc = skb_cow_head(skb, 0);
+	if (unlikely(rc < 0))
+		return rc;
+
+	vhdr = (struct vlan_ethhdr *)skb->data;
+	vhdr->h_vlan_TCI |= cpu_to_be16((skb->priority << VLAN_PRIO_SHIFT)
+					 & VLAN_PRIO_MASK);
 
 	skb->protocol = vlan_get_protocol(skb);
 	return 0;
 }
 
-static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv,
-			  int size, dma_addr_t dma, int frag_end,
-			  enum hns_desc_type type)
+static int hns3_fill_skb_desc(struct hns3_enet_ring *ring,
+			      struct sk_buff *skb, struct hns3_desc *desc)
 {
-	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
-	struct hns3_desc *desc = &ring->desc[ring->next_to_use];
 	u32 ol_type_vlan_len_msec = 0;
-	u16 bdtp_fe_sc_vld_ra_ri = 0;
 	u32 type_cs_vlan_tso = 0;
-	struct sk_buff *skb;
+	u32 paylen = skb->len;
 	u16 inner_vtag = 0;
 	u16 out_vtag = 0;
-	u32 paylen = 0;
 	u16 mss = 0;
-	u8 ol4_proto;
-	u8 il4_proto;
 	int ret;
 
-	/* The txbd's baseinfo of DESC_TYPE_PAGE & DESC_TYPE_SKB */
+	ret = hns3_handle_vtags(ring, skb);
+	if (unlikely(ret < 0)) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.tx_vlan_err++;
+		u64_stats_update_end(&ring->syncp);
+		return ret;
+	} else if (ret == HNS3_INNER_VLAN_TAG) {
+		inner_vtag = skb_vlan_tag_get(skb);
+		inner_vtag |= (skb->priority << VLAN_PRIO_SHIFT) &
+				VLAN_PRIO_MASK;
+		hns3_set_field(type_cs_vlan_tso, HNS3_TXD_VLAN_B, 1);
+	} else if (ret == HNS3_OUTER_VLAN_TAG) {
+		out_vtag = skb_vlan_tag_get(skb);
+		out_vtag |= (skb->priority << VLAN_PRIO_SHIFT) &
+				VLAN_PRIO_MASK;
+		hns3_set_field(ol_type_vlan_len_msec, HNS3_TXD_OVLAN_B,
+			       1);
+	}
+
+	if (skb->ip_summed == CHECKSUM_PARTIAL) {
+		u8 ol4_proto, il4_proto;
+
+		skb_reset_mac_len(skb);
+
+		ret = hns3_get_l4_protocol(skb, &ol4_proto, &il4_proto);
+		if (unlikely(ret < 0)) {
+			u64_stats_update_begin(&ring->syncp);
+			ring->stats.tx_l4_proto_err++;
+			u64_stats_update_end(&ring->syncp);
+			return ret;
+		}
+
+		ret = hns3_set_l2l3l4(skb, ol4_proto, il4_proto,
+				      &type_cs_vlan_tso,
+				      &ol_type_vlan_len_msec);
+		if (unlikely(ret < 0)) {
+			u64_stats_update_begin(&ring->syncp);
+			ring->stats.tx_l2l3l4_err++;
+			u64_stats_update_end(&ring->syncp);
+			return ret;
+		}
+
+		ret = hns3_set_tso(skb, &paylen, &mss,
+				   &type_cs_vlan_tso);
+		if (unlikely(ret < 0)) {
+			u64_stats_update_begin(&ring->syncp);
+			ring->stats.tx_tso_err++;
+			u64_stats_update_end(&ring->syncp);
+			return ret;
+		}
+	}
+
+	/* Set txbd */
+	desc->tx.ol_type_vlan_len_msec =
+		cpu_to_le32(ol_type_vlan_len_msec);
+	desc->tx.type_cs_vlan_tso_len = cpu_to_le32(type_cs_vlan_tso);
+	desc->tx.paylen = cpu_to_le32(paylen);
+	desc->tx.mss = cpu_to_le16(mss);
+	desc->tx.vlan_tag = cpu_to_le16(inner_vtag);
+	desc->tx.outer_vlan_tag = cpu_to_le16(out_vtag);
+
+	return 0;
+}
+
+static int hns3_fill_desc(struct hns3_enet_ring *ring, void *priv,
+			  unsigned int size, enum hns_desc_type type)
+{
+#define HNS3_LIKELY_BD_NUM	1
+
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_use];
+	struct hns3_desc *desc = &ring->desc[ring->next_to_use];
+	struct device *dev = ring_to_dev(ring);
+	skb_frag_t *frag;
+	unsigned int frag_buf_num;
+	int k, sizeoflast;
+	dma_addr_t dma;
+
+	if (type == DESC_TYPE_FRAGLIST_SKB ||
+	    type == DESC_TYPE_SKB) {
+		struct sk_buff *skb = (struct sk_buff *)priv;
+
+		dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE);
+	} else {
+		frag = (skb_frag_t *)priv;
+		dma = skb_frag_dma_map(dev, frag, 0, size, DMA_TO_DEVICE);
+	}
+
+	if (unlikely(dma_mapping_error(dev, dma))) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.sw_err_cnt++;
+		u64_stats_update_end(&ring->syncp);
+		return -ENOMEM;
+	}
+
 	desc_cb->priv = priv;
 	desc_cb->length = size;
 	desc_cb->dma = dma;
 	desc_cb->type = type;
 
-	/* now, fill the descriptor */
-	desc->addr = cpu_to_le64(dma);
-	desc->tx.send_size = cpu_to_le16((u16)size);
-	hns3_set_txbd_baseinfo(&bdtp_fe_sc_vld_ra_ri, frag_end);
-	desc->tx.bdtp_fe_sc_vld_ra_ri = cpu_to_le16(bdtp_fe_sc_vld_ra_ri);
+	if (likely(size <= HNS3_MAX_BD_SIZE)) {
+		desc->addr = cpu_to_le64(dma);
+		desc->tx.send_size = cpu_to_le16(size);
+		desc->tx.bdtp_fe_sc_vld_ra_ri =
+			cpu_to_le16(BIT(HNS3_TXD_VLD_B));
 
-	if (type == DESC_TYPE_SKB) {
-		skb = (struct sk_buff *)priv;
-		paylen = skb->len;
-
-		ret = hns3_fill_desc_vtags(skb, ring, &type_cs_vlan_tso,
-					   &ol_type_vlan_len_msec,
-					   &inner_vtag, &out_vtag);
-		if (unlikely(ret))
-			return ret;
-
-		if (skb->ip_summed == CHECKSUM_PARTIAL) {
-			skb_reset_mac_len(skb);
-
-			ret = hns3_get_l4_protocol(skb, &ol4_proto, &il4_proto);
-			if (ret)
-				return ret;
-			hns3_set_l2l3l4_len(skb, ol4_proto, il4_proto,
-					    &type_cs_vlan_tso,
-					    &ol_type_vlan_len_msec);
-			ret = hns3_set_l3l4_type_csum(skb, ol4_proto, il4_proto,
-						      &type_cs_vlan_tso,
-						      &ol_type_vlan_len_msec);
-			if (ret)
-				return ret;
-
-			ret = hns3_set_tso(skb, &paylen, &mss,
-					   &type_cs_vlan_tso);
-			if (ret)
-				return ret;
-		}
-
-		/* Set txbd */
-		desc->tx.ol_type_vlan_len_msec =
-			cpu_to_le32(ol_type_vlan_len_msec);
-		desc->tx.type_cs_vlan_tso_len =
-			cpu_to_le32(type_cs_vlan_tso);
-		desc->tx.paylen = cpu_to_le32(paylen);
-		desc->tx.mss = cpu_to_le16(mss);
-		desc->tx.vlan_tag = cpu_to_le16(inner_vtag);
-		desc->tx.outer_vlan_tag = cpu_to_le16(out_vtag);
+		trace_hns3_tx_desc(ring, ring->next_to_use);
+		ring_ptr_move_fw(ring, next_to_use);
+		return HNS3_LIKELY_BD_NUM;
 	}
 
-	/* move ring pointer to next.*/
-	ring_ptr_move_fw(ring, next_to_use);
-
-	return 0;
-}
-
-static int hns3_fill_desc_tso(struct hns3_enet_ring *ring, void *priv,
-			      int size, dma_addr_t dma, int frag_end,
-			      enum hns_desc_type type)
-{
-	unsigned int frag_buf_num;
-	unsigned int k;
-	int sizeoflast;
-	int ret;
-
-	frag_buf_num = (size + HNS3_MAX_BD_SIZE - 1) / HNS3_MAX_BD_SIZE;
+	frag_buf_num = hns3_tx_bd_count(size);
 	sizeoflast = size % HNS3_MAX_BD_SIZE;
 	sizeoflast = sizeoflast ? sizeoflast : HNS3_MAX_BD_SIZE;
 
-	/* When the frag size is bigger than hardware, split this frag */
+	/* When frag size is bigger than hardware limit, split this frag */
 	for (k = 0; k < frag_buf_num; k++) {
-		ret = hns3_fill_desc(ring, priv,
-				     (k == frag_buf_num - 1) ?
-				sizeoflast : HNS3_MAX_BD_SIZE,
-				dma + HNS3_MAX_BD_SIZE * k,
-				frag_end && (k == frag_buf_num - 1) ? 1 : 0,
-				(type == DESC_TYPE_SKB && !k) ?
-					DESC_TYPE_SKB : DESC_TYPE_PAGE);
-		if (ret)
-			return ret;
+		/* now, fill the descriptor */
+		desc->addr = cpu_to_le64(dma + HNS3_MAX_BD_SIZE * k);
+		desc->tx.send_size = cpu_to_le16((k == frag_buf_num - 1) ?
+				     (u16)sizeoflast : (u16)HNS3_MAX_BD_SIZE);
+		desc->tx.bdtp_fe_sc_vld_ra_ri =
+				cpu_to_le16(BIT(HNS3_TXD_VLD_B));
+
+		trace_hns3_tx_desc(ring, ring->next_to_use);
+		/* move ring pointer to next */
+		ring_ptr_move_fw(ring, next_to_use);
+
+		desc = &ring->desc[ring->next_to_use];
 	}
 
-	return 0;
+	return frag_buf_num;
 }
 
-static int hns3_nic_maybe_stop_tso(struct sk_buff **out_skb, int *bnum,
-				   struct hns3_enet_ring *ring)
+static unsigned int hns3_skb_bd_num(struct sk_buff *skb, unsigned int *bd_size,
+				    unsigned int bd_num)
 {
-	struct sk_buff *skb = *out_skb;
-	struct skb_frag_struct *frag;
-	int bdnum_for_frag;
-	int frag_num;
-	int buf_num;
-	int size;
+	unsigned int size;
 	int i;
 
 	size = skb_headlen(skb);
-	buf_num = (size + HNS3_MAX_BD_SIZE - 1) / HNS3_MAX_BD_SIZE;
+	while (size > HNS3_MAX_BD_SIZE) {
+		bd_size[bd_num++] = HNS3_MAX_BD_SIZE;
+		size -= HNS3_MAX_BD_SIZE;
 
-	frag_num = skb_shinfo(skb)->nr_frags;
-	for (i = 0; i < frag_num; i++) {
-		frag = &skb_shinfo(skb)->frags[i];
-		size = skb_frag_size(frag);
-		bdnum_for_frag =
-			(size + HNS3_MAX_BD_SIZE - 1) / HNS3_MAX_BD_SIZE;
-		if (bdnum_for_frag > HNS3_MAX_BD_PER_FRAG)
-			return -ENOMEM;
-
-		buf_num += bdnum_for_frag;
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
 	}
 
-	if (buf_num > ring_space(ring))
-		return -EBUSY;
+	if (size) {
+		bd_size[bd_num++] = size;
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
 
-	*bnum = buf_num;
-	return 0;
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+		size = skb_frag_size(frag);
+		if (!size)
+			continue;
+
+		while (size > HNS3_MAX_BD_SIZE) {
+			bd_size[bd_num++] = HNS3_MAX_BD_SIZE;
+			size -= HNS3_MAX_BD_SIZE;
+
+			if (bd_num > HNS3_MAX_TSO_BD_NUM)
+				return bd_num;
+		}
+
+		bd_size[bd_num++] = size;
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	return bd_num;
 }
 
-static int hns3_nic_maybe_stop_tx(struct sk_buff **out_skb, int *bnum,
-				  struct hns3_enet_ring *ring)
+static unsigned int hns3_tx_bd_num(struct sk_buff *skb, unsigned int *bd_size,
+				   u8 max_non_tso_bd_num, unsigned int bd_num,
+				   unsigned int recursion_level)
 {
-	struct sk_buff *skb = *out_skb;
-	int buf_num;
+#define HNS3_MAX_RECURSION_LEVEL	24
 
-	/* No. of segments (plus a header) */
-	buf_num = skb_shinfo(skb)->nr_frags + 1;
+	struct sk_buff *frag_skb;
 
-	if (buf_num > ring_space(ring))
-		return -EBUSY;
+	/* If the total len is within the max bd limit */
+	if (likely(skb->len <= HNS3_MAX_BD_SIZE && !recursion_level &&
+		   !skb_has_frag_list(skb) &&
+		   skb_shinfo(skb)->nr_frags < max_non_tso_bd_num))
+		return skb_shinfo(skb)->nr_frags + 1U;
 
-	*bnum = buf_num;
+	if (unlikely(recursion_level >= HNS3_MAX_RECURSION_LEVEL))
+		return UINT_MAX;
+
+	bd_num = hns3_skb_bd_num(skb, bd_size, bd_num);
+
+	if (!skb_has_frag_list(skb) || bd_num > HNS3_MAX_TSO_BD_NUM)
+		return bd_num;
+
+	skb_walk_frags(skb, frag_skb) {
+		bd_num = hns3_tx_bd_num(frag_skb, bd_size, max_non_tso_bd_num,
+					bd_num, recursion_level + 1);
+		if (bd_num > HNS3_MAX_TSO_BD_NUM)
+			return bd_num;
+	}
+
+	return bd_num;
+}
+
+static unsigned int hns3_gso_hdr_len(struct sk_buff *skb)
+{
+	if (!skb->encapsulation)
+		return skb_transport_offset(skb) + tcp_hdrlen(skb);
+
+	return skb_inner_transport_offset(skb) + inner_tcp_hdrlen(skb);
+}
+
+/* HW need every continuous max_non_tso_bd_num buffer data to be larger
+ * than MSS, we simplify it by ensuring skb_headlen + the first continuous
+ * max_non_tso_bd_num - 1 frags to be larger than gso header len + mss,
+ * and the remaining continuous max_non_tso_bd_num - 1 frags to be larger
+ * than MSS except the last max_non_tso_bd_num - 1 frags.
+ */
+static bool hns3_skb_need_linearized(struct sk_buff *skb, unsigned int *bd_size,
+				     unsigned int bd_num, u8 max_non_tso_bd_num)
+{
+	unsigned int tot_len = 0;
+	int i;
+
+	for (i = 0; i < max_non_tso_bd_num - 1U; i++)
+		tot_len += bd_size[i];
+
+	/* ensure the first max_non_tso_bd_num frags is greater than
+	 * mss + header
+	 */
+	if (tot_len + bd_size[max_non_tso_bd_num - 1U] <
+	    skb_shinfo(skb)->gso_size + hns3_gso_hdr_len(skb))
+		return true;
+
+	/* ensure every continuous max_non_tso_bd_num - 1 buffer is greater
+	 * than mss except the last one.
+	 */
+	for (i = 0; i < bd_num - max_non_tso_bd_num; i++) {
+		tot_len -= bd_size[i];
+		tot_len += bd_size[i + max_non_tso_bd_num - 1U];
+
+		if (tot_len < skb_shinfo(skb)->gso_size)
+			return true;
+	}
+
+	return false;
+}
+
+void hns3_shinfo_pack(struct skb_shared_info *shinfo, __u32 *size)
+{
+	int i;
+
+	for (i = 0; i < MAX_SKB_FRAGS; i++)
+		size[i] = skb_frag_size(&shinfo->frags[i]);
+}
+
+static int hns3_skb_linearize(struct hns3_enet_ring *ring,
+			      struct sk_buff *skb,
+			      unsigned int bd_num)
+{
+	/* 'bd_num == UINT_MAX' means the skb' fraglist has a
+	 * recursion level of over HNS3_MAX_RECURSION_LEVEL.
+	 */
+	if (bd_num == UINT_MAX) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.over_max_recursion++;
+		u64_stats_update_end(&ring->syncp);
+		return -ENOMEM;
+	}
+
+	/* The skb->len has exceeded the hw limitation, linearization
+	 * will not help.
+	 */
+	if (skb->len > HNS3_MAX_TSO_SIZE ||
+	    (!skb_is_gso(skb) && skb->len > HNS3_MAX_NON_TSO_SIZE)) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.hw_limitation++;
+		u64_stats_update_end(&ring->syncp);
+		return -ENOMEM;
+	}
+
+	if (__skb_linearize(skb)) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.sw_err_cnt++;
+		u64_stats_update_end(&ring->syncp);
+		return -ENOMEM;
+	}
 
 	return 0;
 }
 
-static void hns_nic_dma_unmap(struct hns3_enet_ring *ring, int next_to_use_orig)
+static int hns3_nic_maybe_stop_tx(struct hns3_enet_ring *ring,
+				  struct net_device *netdev,
+				  struct sk_buff *skb)
+{
+	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	u8 max_non_tso_bd_num = priv->max_non_tso_bd_num;
+	unsigned int bd_size[HNS3_MAX_TSO_BD_NUM + 1U];
+	unsigned int bd_num;
+
+	bd_num = hns3_tx_bd_num(skb, bd_size, max_non_tso_bd_num, 0, 0);
+	if (unlikely(bd_num > max_non_tso_bd_num)) {
+		if (bd_num <= HNS3_MAX_TSO_BD_NUM && skb_is_gso(skb) &&
+		    !hns3_skb_need_linearized(skb, bd_size, bd_num,
+					      max_non_tso_bd_num)) {
+			trace_hns3_over_max_bd(skb);
+			goto out;
+		}
+
+		if (hns3_skb_linearize(ring, skb, bd_num))
+			return -ENOMEM;
+
+		bd_num = hns3_tx_bd_count(skb->len);
+
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.tx_copy++;
+		u64_stats_update_end(&ring->syncp);
+	}
+
+out:
+	if (likely(ring_space(ring) >= bd_num))
+		return bd_num;
+
+	netif_stop_subqueue(netdev, ring->queue_index);
+	smp_mb(); /* Memory barrier before checking ring_space */
+
+	/* Start queue in case hns3_clean_tx_ring has just made room
+	 * available and has not seen the queue stopped state performed
+	 * by netif_stop_subqueue above.
+	 */
+	if (ring_space(ring) >= bd_num && netif_carrier_ok(netdev) &&
+	    !test_bit(HNS3_NIC_STATE_DOWN, &priv->state)) {
+		netif_start_subqueue(netdev, ring->queue_index);
+		return bd_num;
+	}
+
+	u64_stats_update_begin(&ring->syncp);
+	ring->stats.tx_busy++;
+	u64_stats_update_end(&ring->syncp);
+
+	return -EBUSY;
+}
+
+static void hns3_clear_desc(struct hns3_enet_ring *ring, int next_to_use_orig)
 {
 	struct device *dev = ring_to_dev(ring);
 	unsigned int i;
 
 	for (i = 0; i < ring->desc_num; i++) {
+		struct hns3_desc *desc = &ring->desc[ring->next_to_use];
+
+		memset(desc, 0, sizeof(*desc));
+
 		/* check if this is where we started */
 		if (ring->next_to_use == next_to_use_orig)
 			break;
 
+		/* rollback one */
+		ring_ptr_move_bw(ring, next_to_use);
+
+		if (!ring->desc_cb[ring->next_to_use].dma)
+			continue;
+
 		/* unmap the descriptor dma address */
-		if (ring->desc_cb[ring->next_to_use].type == DESC_TYPE_SKB)
+		if (ring->desc_cb[ring->next_to_use].type == DESC_TYPE_SKB ||
+		    ring->desc_cb[ring->next_to_use].type ==
+		    DESC_TYPE_FRAGLIST_SKB)
 			dma_unmap_single(dev,
 					 ring->desc_cb[ring->next_to_use].dma,
 					ring->desc_cb[ring->next_to_use].length,
 					DMA_TO_DEVICE);
-		else
+		else if (ring->desc_cb[ring->next_to_use].length)
 			dma_unmap_page(dev,
 				       ring->desc_cb[ring->next_to_use].dma,
 				       ring->desc_cb[ring->next_to_use].length,
 				       DMA_TO_DEVICE);
 
-		/* rollback one */
-		ring_ptr_move_bw(ring, next_to_use);
+		ring->desc_cb[ring->next_to_use].length = 0;
+		ring->desc_cb[ring->next_to_use].dma = 0;
+		ring->desc_cb[ring->next_to_use].type = DESC_TYPE_UNKNOWN;
 	}
+}
+
+static int hns3_fill_skb_to_desc(struct hns3_enet_ring *ring,
+				 struct sk_buff *skb, enum hns_desc_type type)
+{
+	unsigned int size = skb_headlen(skb);
+	struct sk_buff *frag_skb;
+	int i, ret, bd_num = 0;
+
+	if (size) {
+		ret = hns3_fill_desc(ring, skb, size, type);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+
+		size = skb_frag_size(frag);
+		if (!size)
+			continue;
+
+		ret = hns3_fill_desc(ring, frag, size, DESC_TYPE_PAGE);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	skb_walk_frags(skb, frag_skb) {
+		ret = hns3_fill_skb_to_desc(ring, frag_skb,
+					    DESC_TYPE_FRAGLIST_SKB);
+		if (unlikely(ret < 0))
+			return ret;
+
+		bd_num += ret;
+	}
+
+	return bd_num;
+}
+
+static void hns3_tx_doorbell(struct hns3_enet_ring *ring, int num,
+			     bool doorbell)
+{
+	ring->pending_buf += num;
+
+	if (!doorbell) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.tx_more++;
+		u64_stats_update_end(&ring->syncp);
+		return;
+	}
+
+	if (!ring->pending_buf)
+		return;
+
+	writel(ring->pending_buf,
+	       ring->tqp->io_base + HNS3_RING_TX_RING_TAIL_REG);
+	ring->pending_buf = 0;
+	WRITE_ONCE(ring->last_to_use, ring->next_to_use);
 }
 
 netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
-	struct hns3_nic_ring_data *ring_data =
-		&tx_ring_data(priv, skb->queue_mapping);
-	struct hns3_enet_ring *ring = ring_data->ring;
-	struct device *dev = priv->dev;
+	struct hns3_enet_ring *ring = &priv->ring[skb->queue_mapping];
 	struct netdev_queue *dev_queue;
-	struct skb_frag_struct *frag;
-	int next_to_use_head;
-	int next_to_use_frag;
-	dma_addr_t dma;
-	int buf_num;
-	int seg_num;
-	int size;
+	int pre_ntu, next_to_use_head;
+	bool doorbell;
 	int ret;
-	int i;
+
+	/* Hardware can only handle short frames above 32 bytes */
+	if (skb_put_padto(skb, HNS3_MIN_TX_LEN)) {
+		hns3_tx_doorbell(ring, 0, !netdev_xmit_more());
+		return NETDEV_TX_OK;
+	}
 
 	/* Prefetch the data used later */
 	prefetch(skb->data);
 
-	switch (priv->ops.maybe_stop_tx(&skb, &buf_num, ring)) {
-	case -EBUSY:
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.tx_busy++;
-		u64_stats_update_end(&ring->syncp);
+	ret = hns3_nic_maybe_stop_tx(ring, netdev, skb);
+	if (unlikely(ret <= 0)) {
+		if (ret == -EBUSY) {
+			hns3_tx_doorbell(ring, 0, true);
+			return NETDEV_TX_BUSY;
+		}
 
-		goto out_net_tx_busy;
-	case -ENOMEM:
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.sw_err_cnt++;
-		u64_stats_update_end(&ring->syncp);
-		netdev_err(netdev, "no memory to xmit!\n");
-
+		hns3_rl_err(netdev, "xmit error: %d!\n", ret);
 		goto out_err_tx_ok;
-	default:
-		break;
 	}
-
-	/* No. of segments (plus a header) */
-	seg_num = skb_shinfo(skb)->nr_frags + 1;
-	/* Fill the first part */
-	size = skb_headlen(skb);
 
 	next_to_use_head = ring->next_to_use;
 
-	dma = dma_map_single(dev, skb->data, size, DMA_TO_DEVICE);
-	if (dma_mapping_error(dev, dma)) {
-		netdev_err(netdev, "TX head DMA map failed\n");
-		ring->stats.sw_err_cnt++;
-		goto out_err_tx_ok;
-	}
+	ret = hns3_fill_skb_desc(ring, skb, &ring->desc[ring->next_to_use]);
+	if (unlikely(ret < 0))
+		goto fill_err;
 
-	ret = priv->ops.fill_desc(ring, skb, size, dma, seg_num == 1 ? 1 : 0,
-			   DESC_TYPE_SKB);
-	if (ret)
-		goto head_dma_map_err;
+	/* 'ret < 0' means filling error, 'ret == 0' means skb->len is
+	 * zero, which is unlikely, and 'ret > 0' means how many tx desc
+	 * need to be notified to the hw.
+	 */
+	ret = hns3_fill_skb_to_desc(ring, skb, DESC_TYPE_SKB);
+	if (unlikely(ret <= 0))
+		goto fill_err;
 
-	next_to_use_frag = ring->next_to_use;
-	/* Fill the fragments */
-	for (i = 1; i < seg_num; i++) {
-		frag = &skb_shinfo(skb)->frags[i - 1];
-		size = skb_frag_size(frag);
-		dma = skb_frag_dma_map(dev, frag, 0, size, DMA_TO_DEVICE);
-		if (dma_mapping_error(dev, dma)) {
-			netdev_err(netdev, "TX frag(%d) DMA map failed\n", i);
-			ring->stats.sw_err_cnt++;
-			goto frag_dma_map_err;
-		}
-		ret = priv->ops.fill_desc(ring, skb_frag_page(frag), size, dma,
-				    seg_num - 1 == i ? 1 : 0,
-				    DESC_TYPE_PAGE);
-
-		if (ret)
-			goto frag_dma_map_err;
-	}
+	pre_ntu = ring->next_to_use ? (ring->next_to_use - 1) :
+					(ring->desc_num - 1);
+	ring->desc[pre_ntu].tx.bdtp_fe_sc_vld_ra_ri |=
+				cpu_to_le16(BIT(HNS3_TXD_FE_B));
+	trace_hns3_tx_desc(ring, pre_ntu);
 
 	/* Complete translate all packets */
-	dev_queue = netdev_get_tx_queue(netdev, ring_data->queue_index);
-	netdev_tx_sent_queue(dev_queue, skb->len);
-
-	wmb(); /* Commit all data before submit */
-
-	hnae3_queue_xmit(ring->tqp, buf_num);
+	dev_queue = netdev_get_tx_queue(netdev, ring->queue_index);
+	doorbell = __netdev_tx_sent_queue(dev_queue, skb->len,
+					  netdev_xmit_more());
+	hns3_tx_doorbell(ring, ret, doorbell);
 
 	return NETDEV_TX_OK;
 
-frag_dma_map_err:
-	hns_nic_dma_unmap(ring, next_to_use_frag);
-
-head_dma_map_err:
-	hns_nic_dma_unmap(ring, next_to_use_head);
+fill_err:
+	hns3_clear_desc(ring, next_to_use_head);
 
 out_err_tx_ok:
 	dev_kfree_skb_any(skb);
+	hns3_tx_doorbell(ring, 0, !netdev_xmit_more());
 	return NETDEV_TX_OK;
-
-out_net_tx_busy:
-	netif_stop_subqueue(netdev, ring_data->queue_index);
-	smp_mb(); /* Commit all data before submit */
-
-	return NETDEV_TX_BUSY;
 }
 
 static int hns3_nic_net_set_mac_address(struct net_device *netdev, void *p)
@@ -1194,6 +1551,16 @@
 		return 0;
 	}
 
+	/* For VF device, if there is a perm_addr, then the user will not
+	 * be allowed to change the address.
+	 */
+	if (!hns3_is_phys_func(h->pdev) &&
+	    !is_zero_ether_addr(netdev->perm_addr)) {
+		netdev_err(netdev, "has permanent MAC %pM, user MAC %pM not allow\n",
+			   netdev->perm_addr, mac_addr->sa_data);
+		return -EPERM;
+	}
+
 	ret = h->ae_algo->ops->set_mac_addr(h, mac_addr->sa_data, false);
 	if (ret) {
 		netdev_err(netdev, "set_mac_address fail, ret=%d!\n", ret);
@@ -1205,45 +1572,82 @@
 	return 0;
 }
 
+static int hns3_nic_do_ioctl(struct net_device *netdev,
+			     struct ifreq *ifr, int cmd)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!netif_running(netdev))
+		return -EINVAL;
+
+	if (!h->ae_algo->ops->do_ioctl)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->do_ioctl(h, ifr, cmd);
+}
+
 static int hns3_nic_set_features(struct net_device *netdev,
 				 netdev_features_t features)
 {
 	netdev_features_t changed = netdev->features ^ features;
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	struct hnae3_handle *h = priv->ae_handle;
+	bool enable;
 	int ret;
 
-	if (changed & (NETIF_F_TSO | NETIF_F_TSO6)) {
-		if (features & (NETIF_F_TSO | NETIF_F_TSO6)) {
-			priv->ops.fill_desc = hns3_fill_desc_tso;
-			priv->ops.maybe_stop_tx = hns3_nic_maybe_stop_tso;
-		} else {
-			priv->ops.fill_desc = hns3_fill_desc;
-			priv->ops.maybe_stop_tx = hns3_nic_maybe_stop_tx;
-		}
-	}
-
-	if ((changed & NETIF_F_HW_VLAN_CTAG_FILTER) &&
-	    h->ae_algo->ops->enable_vlan_filter) {
-		if (features & NETIF_F_HW_VLAN_CTAG_FILTER)
-			h->ae_algo->ops->enable_vlan_filter(h, true);
-		else
-			h->ae_algo->ops->enable_vlan_filter(h, false);
-	}
-
-	if ((changed & NETIF_F_HW_VLAN_CTAG_RX) &&
-	    h->ae_algo->ops->enable_hw_strip_rxvtag) {
-		if (features & NETIF_F_HW_VLAN_CTAG_RX)
-			ret = h->ae_algo->ops->enable_hw_strip_rxvtag(h, true);
-		else
-			ret = h->ae_algo->ops->enable_hw_strip_rxvtag(h, false);
-
+	if (changed & (NETIF_F_GRO_HW) && h->ae_algo->ops->set_gro_en) {
+		enable = !!(features & NETIF_F_GRO_HW);
+		ret = h->ae_algo->ops->set_gro_en(h, enable);
 		if (ret)
 			return ret;
 	}
 
+	if ((changed & NETIF_F_HW_VLAN_CTAG_RX) &&
+	    h->ae_algo->ops->enable_hw_strip_rxvtag) {
+		enable = !!(features & NETIF_F_HW_VLAN_CTAG_RX);
+		ret = h->ae_algo->ops->enable_hw_strip_rxvtag(h, enable);
+		if (ret)
+			return ret;
+	}
+
+	if ((changed & NETIF_F_NTUPLE) && h->ae_algo->ops->enable_fd) {
+		enable = !!(features & NETIF_F_NTUPLE);
+		h->ae_algo->ops->enable_fd(h, enable);
+	}
+
 	netdev->features = features;
 	return 0;
+}
+
+static netdev_features_t hns3_features_check(struct sk_buff *skb,
+					     struct net_device *dev,
+					     netdev_features_t features)
+{
+#define HNS3_MAX_HDR_LEN	480U
+#define HNS3_MAX_L4_HDR_LEN	60U
+
+	size_t len;
+
+	if (skb->ip_summed != CHECKSUM_PARTIAL)
+		return features;
+
+	if (skb->encapsulation)
+		len = skb_inner_transport_header(skb) - skb->data;
+	else
+		len = skb_transport_header(skb) - skb->data;
+
+	/* Assume L4 is 60 byte as TCP is the only protocol with a
+	 * a flexible value, and it's max len is 60 bytes.
+	 */
+	len += HNS3_MAX_L4_HDR_LEN;
+
+	/* Hardware only supports checksum on the skb with a max header
+	 * len of 480 bytes.
+	 */
+	if (len > HNS3_MAX_HDR_LEN)
+		features &= ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
+
+	return features;
 }
 
 static void hns3_nic_get_stats64(struct net_device *netdev,
@@ -1253,7 +1657,12 @@
 	int queue_num = priv->ae_handle->kinfo.num_tqps;
 	struct hnae3_handle *handle = priv->ae_handle;
 	struct hns3_enet_ring *ring;
+	u64 rx_length_errors = 0;
+	u64 rx_crc_errors = 0;
+	u64 rx_multicast = 0;
 	unsigned int start;
+	u64 tx_errors = 0;
+	u64 rx_errors = 0;
 	unsigned int idx;
 	u64 tx_bytes = 0;
 	u64 rx_bytes = 0;
@@ -1269,24 +1678,39 @@
 
 	for (idx = 0; idx < queue_num; idx++) {
 		/* fetch the tx stats */
-		ring = priv->ring_data[idx].ring;
+		ring = &priv->ring[idx];
 		do {
 			start = u64_stats_fetch_begin_irq(&ring->syncp);
 			tx_bytes += ring->stats.tx_bytes;
 			tx_pkts += ring->stats.tx_pkts;
-			tx_drop += ring->stats.tx_busy;
 			tx_drop += ring->stats.sw_err_cnt;
+			tx_drop += ring->stats.tx_vlan_err;
+			tx_drop += ring->stats.tx_l4_proto_err;
+			tx_drop += ring->stats.tx_l2l3l4_err;
+			tx_drop += ring->stats.tx_tso_err;
+			tx_drop += ring->stats.over_max_recursion;
+			tx_drop += ring->stats.hw_limitation;
+			tx_errors += ring->stats.sw_err_cnt;
+			tx_errors += ring->stats.tx_vlan_err;
+			tx_errors += ring->stats.tx_l4_proto_err;
+			tx_errors += ring->stats.tx_l2l3l4_err;
+			tx_errors += ring->stats.tx_tso_err;
+			tx_errors += ring->stats.over_max_recursion;
+			tx_errors += ring->stats.hw_limitation;
 		} while (u64_stats_fetch_retry_irq(&ring->syncp, start));
 
 		/* fetch the rx stats */
-		ring = priv->ring_data[idx + queue_num].ring;
+		ring = &priv->ring[idx + queue_num];
 		do {
 			start = u64_stats_fetch_begin_irq(&ring->syncp);
 			rx_bytes += ring->stats.rx_bytes;
 			rx_pkts += ring->stats.rx_pkts;
-			rx_drop += ring->stats.non_vld_descs;
-			rx_drop += ring->stats.err_pkt_len;
 			rx_drop += ring->stats.l2_err;
+			rx_errors += ring->stats.l2_err;
+			rx_errors += ring->stats.l3l4_csum_err;
+			rx_crc_errors += ring->stats.l2_err;
+			rx_multicast += ring->stats.rx_multicast;
+			rx_length_errors += ring->stats.err_pkt_len;
 		} while (u64_stats_fetch_retry_irq(&ring->syncp, start));
 	}
 
@@ -1295,15 +1719,15 @@
 	stats->rx_bytes = rx_bytes;
 	stats->rx_packets = rx_pkts;
 
-	stats->rx_errors = netdev->stats.rx_errors;
-	stats->multicast = netdev->stats.multicast;
-	stats->rx_length_errors = netdev->stats.rx_length_errors;
-	stats->rx_crc_errors = netdev->stats.rx_crc_errors;
+	stats->rx_errors = rx_errors;
+	stats->multicast = rx_multicast;
+	stats->rx_length_errors = rx_length_errors;
+	stats->rx_crc_errors = rx_crc_errors;
 	stats->rx_missed_errors = netdev->stats.rx_missed_errors;
 
-	stats->tx_errors = netdev->stats.tx_errors;
-	stats->rx_dropped = rx_drop + netdev->stats.rx_dropped;
-	stats->tx_dropped = tx_drop + netdev->stats.tx_dropped;
+	stats->tx_errors = tx_errors;
+	stats->rx_dropped = rx_drop;
+	stats->tx_dropped = tx_drop;
 	stats->collisions = netdev->stats.collisions;
 	stats->rx_over_errors = netdev->stats.rx_over_errors;
 	stats->rx_frame_errors = netdev->stats.rx_frame_errors;
@@ -1320,14 +1744,12 @@
 static int hns3_setup_tc(struct net_device *netdev, void *type_data)
 {
 	struct tc_mqprio_qopt_offload *mqprio_qopt = type_data;
-	struct hnae3_handle *h = hns3_get_handle(netdev);
-	struct hnae3_knic_private_info *kinfo = &h->kinfo;
 	u8 *prio_tc = mqprio_qopt->qopt.prio_tc_map;
+	struct hnae3_knic_private_info *kinfo;
 	u8 tc = mqprio_qopt->qopt.num_tc;
 	u16 mode = mqprio_qopt->mode;
 	u8 hw = mqprio_qopt->qopt.hw;
-	bool if_running;
-	int ret;
+	struct hnae3_handle *h;
 
 	if (!((hw == TC_MQPRIO_HW_OFFLOAD_TCS &&
 	       mode == TC_MQPRIO_MODE_CHANNEL) || (!hw && tc == 0)))
@@ -1339,24 +1761,13 @@
 	if (!netdev)
 		return -EINVAL;
 
-	if_running = netif_running(netdev);
-	if (if_running) {
-		hns3_nic_net_stop(netdev);
-		msleep(100);
-	}
+	h = hns3_get_handle(netdev);
+	kinfo = &h->kinfo;
 
-	ret = (kinfo->dcb_ops && kinfo->dcb_ops->setup_tc) ?
-		kinfo->dcb_ops->setup_tc(h, tc, prio_tc) : -EOPNOTSUPP;
-	if (ret)
-		goto out;
+	netif_dbg(h, drv, netdev, "setup tc: num_tc=%u\n", tc);
 
-	ret = hns3_nic_set_real_num_queue(netdev);
-
-out:
-	if (if_running)
-		hns3_nic_net_open(netdev);
-
-	return ret;
+	return (kinfo->dcb_ops && kinfo->dcb_ops->setup_tc) ?
+		kinfo->dcb_ops->setup_tc(h, tc ? tc : 1, prio_tc) : -EOPNOTSUPP;
 }
 
 static int hns3_nic_setup_tc(struct net_device *dev, enum tc_setup_type type,
@@ -1372,14 +1783,10 @@
 				__be16 proto, u16 vid)
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	int ret = -EIO;
 
 	if (h->ae_algo->ops->set_vlan_filter)
 		ret = h->ae_algo->ops->set_vlan_filter(h, proto, vid, false);
-
-	if (!ret)
-		set_bit(vid, priv->active_vlans);
 
 	return ret;
 }
@@ -1388,30 +1795,12 @@
 				 __be16 proto, u16 vid)
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	int ret = -EIO;
 
 	if (h->ae_algo->ops->set_vlan_filter)
 		ret = h->ae_algo->ops->set_vlan_filter(h, proto, vid, true);
 
-	if (!ret)
-		clear_bit(vid, priv->active_vlans);
-
 	return ret;
-}
-
-static void hns3_restore_vlan(struct net_device *netdev)
-{
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
-	u16 vid;
-	int ret;
-
-	for_each_set_bit(vid, priv->active_vlans, VLAN_N_VID) {
-		ret = hns3_vlan_rx_add_vid(netdev, htons(ETH_P_8021Q), vid);
-		if (ret)
-			netdev_warn(netdev, "Restore vlan: %d filter, ret:%d\n",
-				    vid, ret);
-	}
 }
 
 static int hns3_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan,
@@ -1420,27 +1809,53 @@
 	struct hnae3_handle *h = hns3_get_handle(netdev);
 	int ret = -EIO;
 
+	netif_dbg(h, drv, netdev,
+		  "set vf vlan: vf=%d, vlan=%u, qos=%u, vlan_proto=0x%x\n",
+		  vf, vlan, qos, ntohs(vlan_proto));
+
 	if (h->ae_algo->ops->set_vf_vlan_filter)
 		ret = h->ae_algo->ops->set_vf_vlan_filter(h, vf, vlan,
-						   qos, vlan_proto);
+							  qos, vlan_proto);
 
 	return ret;
+}
+
+static int hns3_set_vf_spoofchk(struct net_device *netdev, int vf, bool enable)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
+
+	if (!handle->ae_algo->ops->set_vf_spoofchk)
+		return -EOPNOTSUPP;
+
+	return handle->ae_algo->ops->set_vf_spoofchk(handle, vf, enable);
+}
+
+static int hns3_set_vf_trust(struct net_device *netdev, int vf, bool enable)
+{
+	struct hnae3_handle *handle = hns3_get_handle(netdev);
+
+	if (!handle->ae_algo->ops->set_vf_trust)
+		return -EOPNOTSUPP;
+
+	return handle->ae_algo->ops->set_vf_trust(handle, vf, enable);
 }
 
 static int hns3_nic_change_mtu(struct net_device *netdev, int new_mtu)
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
-	bool if_running = netif_running(netdev);
 	int ret;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
 
 	if (!h->ae_algo->ops->set_mtu)
 		return -EOPNOTSUPP;
 
-	/* if this was called with netdev up then bring netdevice down */
-	if (if_running) {
-		(void)hns3_nic_net_stop(netdev);
-		msleep(100);
-	}
+	netif_dbg(h, drv, netdev,
+		  "change mtu from %u to %d\n", netdev->mtu, new_mtu);
 
 	ret = h->ae_algo->ops->set_mtu(h, new_mtu);
 	if (ret)
@@ -1449,19 +1864,21 @@
 	else
 		netdev->mtu = new_mtu;
 
-	/* if the netdev was running earlier, bring it up again */
-	if (if_running && hns3_nic_net_open(netdev))
-		ret = -EINVAL;
-
 	return ret;
 }
 
 static bool hns3_get_tx_timeo_queue_info(struct net_device *ndev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
-	struct hns3_enet_ring *tx_ring = NULL;
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+	struct hns3_enet_ring *tx_ring;
+	struct napi_struct *napi;
 	int timeout_queue = 0;
 	int hw_head, hw_tail;
+	int fbd_num, fbd_oft;
+	int ebd_num, ebd_oft;
+	int bd_num, bd_err;
+	int ring_en, tc;
 	int i;
 
 	/* Find the stopped queue the same way the stack does */
@@ -1489,26 +1906,68 @@
 		return false;
 	}
 
-	tx_ring = priv->ring_data[timeout_queue].ring;
+	priv->tx_timeout_count++;
+
+	tx_ring = &priv->ring[timeout_queue];
+	napi = &tx_ring->tqp_vector->napi;
+
+	netdev_info(ndev,
+		    "tx_timeout count: %llu, queue id: %d, SW_NTU: 0x%x, SW_NTC: 0x%x, napi state: %lu\n",
+		    priv->tx_timeout_count, timeout_queue, tx_ring->next_to_use,
+		    tx_ring->next_to_clean, napi->state);
+
+	netdev_info(ndev,
+		    "tx_pkts: %llu, tx_bytes: %llu, sw_err_cnt: %llu, tx_pending: %d\n",
+		    tx_ring->stats.tx_pkts, tx_ring->stats.tx_bytes,
+		    tx_ring->stats.sw_err_cnt, tx_ring->pending_buf);
+
+	netdev_info(ndev,
+		    "seg_pkt_cnt: %llu, tx_more: %llu, restart_queue: %llu, tx_busy: %llu\n",
+		    tx_ring->stats.seg_pkt_cnt, tx_ring->stats.tx_more,
+		    tx_ring->stats.restart_queue, tx_ring->stats.tx_busy);
+
+	/* When mac received many pause frames continuous, it's unable to send
+	 * packets, which may cause tx timeout
+	 */
+	if (h->ae_algo->ops->get_mac_stats) {
+		struct hns3_mac_stats mac_stats;
+
+		h->ae_algo->ops->get_mac_stats(h, &mac_stats);
+		netdev_info(ndev, "tx_pause_cnt: %llu, rx_pause_cnt: %llu\n",
+			    mac_stats.tx_pause_cnt, mac_stats.rx_pause_cnt);
+	}
 
 	hw_head = readl_relaxed(tx_ring->tqp->io_base +
 				HNS3_RING_TX_RING_HEAD_REG);
 	hw_tail = readl_relaxed(tx_ring->tqp->io_base +
 				HNS3_RING_TX_RING_TAIL_REG);
+	fbd_num = readl_relaxed(tx_ring->tqp->io_base +
+				HNS3_RING_TX_RING_FBDNUM_REG);
+	fbd_oft = readl_relaxed(tx_ring->tqp->io_base +
+				HNS3_RING_TX_RING_OFFSET_REG);
+	ebd_num = readl_relaxed(tx_ring->tqp->io_base +
+				HNS3_RING_TX_RING_EBDNUM_REG);
+	ebd_oft = readl_relaxed(tx_ring->tqp->io_base +
+				HNS3_RING_TX_RING_EBD_OFFSET_REG);
+	bd_num = readl_relaxed(tx_ring->tqp->io_base +
+			       HNS3_RING_TX_RING_BD_NUM_REG);
+	bd_err = readl_relaxed(tx_ring->tqp->io_base +
+			       HNS3_RING_TX_RING_BD_ERR_REG);
+	ring_en = readl_relaxed(tx_ring->tqp->io_base + HNS3_RING_EN_REG);
+	tc = readl_relaxed(tx_ring->tqp->io_base + HNS3_RING_TX_RING_TC_REG);
+
 	netdev_info(ndev,
-		    "tx_timeout count: %llu, queue id: %d, SW_NTU: 0x%x, SW_NTC: 0x%x, HW_HEAD: 0x%x, HW_TAIL: 0x%x, INT: 0x%x\n",
-		    priv->tx_timeout_count,
-		    timeout_queue,
-		    tx_ring->next_to_use,
-		    tx_ring->next_to_clean,
-		    hw_head,
-		    hw_tail,
+		    "BD_NUM: 0x%x HW_HEAD: 0x%x, HW_TAIL: 0x%x, BD_ERR: 0x%x, INT: 0x%x\n",
+		    bd_num, hw_head, hw_tail, bd_err,
 		    readl(tx_ring->tqp_vector->mask_addr));
+	netdev_info(ndev,
+		    "RING_EN: 0x%x, TC: 0x%x, FBD_NUM: 0x%x FBD_OFT: 0x%x, EBD_NUM: 0x%x, EBD_OFT: 0x%x\n",
+		    ring_en, tc, fbd_num, fbd_oft, ebd_num, ebd_oft);
 
 	return true;
 }
 
-static void hns3_nic_net_timeout(struct net_device *ndev)
+static void hns3_nic_net_timeout(struct net_device *ndev, unsigned int txqueue)
 {
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
 	struct hnae3_handle *h = priv->ae_handle;
@@ -1516,14 +1975,88 @@
 	if (!hns3_get_tx_timeo_queue_info(ndev))
 		return;
 
-	priv->tx_timeout_count++;
-
-	if (time_before(jiffies, (h->last_reset_time + ndev->watchdog_timeo)))
-		return;
-
-	/* request the reset */
+	/* request the reset, and let the hclge to determine
+	 * which reset level should be done
+	 */
 	if (h->ae_algo->ops->reset_event)
-		h->ae_algo->ops->reset_event(h);
+		h->ae_algo->ops->reset_event(h->pdev, h);
+}
+
+#ifdef CONFIG_RFS_ACCEL
+static int hns3_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb,
+			      u16 rxq_index, u32 flow_id)
+{
+	struct hnae3_handle *h = hns3_get_handle(dev);
+	struct flow_keys fkeys;
+
+	if (!h->ae_algo->ops->add_arfs_entry)
+		return -EOPNOTSUPP;
+
+	if (skb->encapsulation)
+		return -EPROTONOSUPPORT;
+
+	if (!skb_flow_dissect_flow_keys(skb, &fkeys, 0))
+		return -EPROTONOSUPPORT;
+
+	if ((fkeys.basic.n_proto != htons(ETH_P_IP) &&
+	     fkeys.basic.n_proto != htons(ETH_P_IPV6)) ||
+	    (fkeys.basic.ip_proto != IPPROTO_TCP &&
+	     fkeys.basic.ip_proto != IPPROTO_UDP))
+		return -EPROTONOSUPPORT;
+
+	return h->ae_algo->ops->add_arfs_entry(h, rxq_index, flow_id, &fkeys);
+}
+#endif
+
+static int hns3_nic_get_vf_config(struct net_device *ndev, int vf,
+				  struct ifla_vf_info *ivf)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->get_vf_config)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->get_vf_config(h, vf, ivf);
+}
+
+static int hns3_nic_set_vf_link_state(struct net_device *ndev, int vf,
+				      int link_state)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->set_vf_link_state)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->set_vf_link_state(h, vf, link_state);
+}
+
+static int hns3_nic_set_vf_rate(struct net_device *ndev, int vf,
+				int min_tx_rate, int max_tx_rate)
+{
+	struct hnae3_handle *h = hns3_get_handle(ndev);
+
+	if (!h->ae_algo->ops->set_vf_rate)
+		return -EOPNOTSUPP;
+
+	return h->ae_algo->ops->set_vf_rate(h, vf, min_tx_rate, max_tx_rate,
+					    false);
+}
+
+static int hns3_nic_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (!h->ae_algo->ops->set_vf_mac)
+		return -EOPNOTSUPP;
+
+	if (is_multicast_ether_addr(mac)) {
+		netdev_err(netdev,
+			   "Invalid MAC:%pM specified. Could not set MAC\n",
+			   mac);
+		return -EINVAL;
+	}
+
+	return h->ae_algo->ops->set_vf_mac(h, vf_id, mac);
 }
 
 static const struct net_device_ops hns3_nic_netdev_ops = {
@@ -1532,17 +2065,28 @@
 	.ndo_start_xmit		= hns3_nic_net_xmit,
 	.ndo_tx_timeout		= hns3_nic_net_timeout,
 	.ndo_set_mac_address	= hns3_nic_net_set_mac_address,
+	.ndo_do_ioctl		= hns3_nic_do_ioctl,
 	.ndo_change_mtu		= hns3_nic_change_mtu,
 	.ndo_set_features	= hns3_nic_set_features,
+	.ndo_features_check	= hns3_features_check,
 	.ndo_get_stats64	= hns3_nic_get_stats64,
 	.ndo_setup_tc		= hns3_nic_setup_tc,
 	.ndo_set_rx_mode	= hns3_nic_set_rx_mode,
 	.ndo_vlan_rx_add_vid	= hns3_vlan_rx_add_vid,
 	.ndo_vlan_rx_kill_vid	= hns3_vlan_rx_kill_vid,
 	.ndo_set_vf_vlan	= hns3_ndo_set_vf_vlan,
+	.ndo_set_vf_spoofchk	= hns3_set_vf_spoofchk,
+	.ndo_set_vf_trust	= hns3_set_vf_trust,
+#ifdef CONFIG_RFS_ACCEL
+	.ndo_rx_flow_steer	= hns3_rx_flow_steer,
+#endif
+	.ndo_get_vf_config	= hns3_nic_get_vf_config,
+	.ndo_set_vf_link_state	= hns3_nic_set_vf_link_state,
+	.ndo_set_vf_rate	= hns3_nic_set_vf_rate,
+	.ndo_set_vf_mac		= hns3_nic_set_vf_mac,
 };
 
-static bool hns3_is_phys_func(struct pci_dev *pdev)
+bool hns3_is_phys_func(struct pci_dev *pdev)
 {
 	u32 dev_id = pdev->device;
 
@@ -1554,12 +2098,13 @@
 	case HNAE3_DEV_ID_50GE_RDMA:
 	case HNAE3_DEV_ID_50GE_RDMA_MACSEC:
 	case HNAE3_DEV_ID_100G_RDMA_MACSEC:
+	case HNAE3_DEV_ID_200G_RDMA:
 		return true;
-	case HNAE3_DEV_ID_100G_VF:
-	case HNAE3_DEV_ID_100G_RDMA_DCB_PFC_VF:
+	case HNAE3_DEV_ID_VF:
+	case HNAE3_DEV_ID_RDMA_DCB_PFC_VF:
 		return false;
 	default:
-		dev_warn(&pdev->dev, "un-recognized pci device-id %d",
+		dev_warn(&pdev->dev, "un-recognized pci device-id %u",
 			 dev_id);
 	}
 
@@ -1596,23 +2141,17 @@
 	struct hnae3_ae_dev *ae_dev;
 	int ret;
 
-	ae_dev = devm_kzalloc(&pdev->dev, sizeof(*ae_dev),
-			      GFP_KERNEL);
-	if (!ae_dev) {
-		ret = -ENOMEM;
-		return ret;
-	}
+	ae_dev = devm_kzalloc(&pdev->dev, sizeof(*ae_dev), GFP_KERNEL);
+	if (!ae_dev)
+		return -ENOMEM;
 
 	ae_dev->pdev = pdev;
 	ae_dev->flag = ent->driver_data;
-	ae_dev->dev_type = HNAE3_DEV_KNIC;
 	pci_set_drvdata(pdev, ae_dev);
 
 	ret = hnae3_register_ae_dev(ae_dev);
-	if (ret) {
-		devm_kfree(&pdev->dev, ae_dev);
+	if (ret)
 		pci_set_drvdata(pdev, NULL);
-	}
 
 	return ret;
 }
@@ -1664,12 +2203,103 @@
 	return 0;
 }
 
+static void hns3_shutdown(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	hnae3_unregister_ae_dev(ae_dev);
+	pci_set_drvdata(pdev, NULL);
+
+	if (system_state == SYSTEM_POWER_OFF)
+		pci_set_power_state(pdev, PCI_D3hot);
+}
+
+static pci_ers_result_t hns3_error_detected(struct pci_dev *pdev,
+					    pci_channel_state_t state)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	pci_ers_result_t ret;
+
+	dev_info(&pdev->dev, "PCI error detected, state(=%d)!!\n", state);
+
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
+	if (!ae_dev || !ae_dev->ops) {
+		dev_err(&pdev->dev,
+			"Can't recover - error happened before device initialized\n");
+		return PCI_ERS_RESULT_NONE;
+	}
+
+	if (ae_dev->ops->handle_hw_ras_error)
+		ret = ae_dev->ops->handle_hw_ras_error(ae_dev);
+	else
+		return PCI_ERS_RESULT_NONE;
+
+	return ret;
+}
+
+static pci_ers_result_t hns3_slot_reset(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	const struct hnae3_ae_ops *ops;
+	enum hnae3_reset_type reset_type;
+	struct device *dev = &pdev->dev;
+
+	if (!ae_dev || !ae_dev->ops)
+		return PCI_ERS_RESULT_NONE;
+
+	ops = ae_dev->ops;
+	/* request the reset */
+	if (ops->reset_event && ops->get_reset_level &&
+	    ops->set_default_reset_request) {
+		if (ae_dev->hw_err_reset_req) {
+			reset_type = ops->get_reset_level(ae_dev,
+						&ae_dev->hw_err_reset_req);
+			ops->set_default_reset_request(ae_dev, reset_type);
+			dev_info(dev, "requesting reset due to PCI error\n");
+			ops->reset_event(pdev, NULL);
+		}
+
+		return PCI_ERS_RESULT_RECOVERED;
+	}
+
+	return PCI_ERS_RESULT_DISCONNECT;
+}
+
+static void hns3_reset_prepare(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	dev_info(&pdev->dev, "FLR prepare\n");
+	if (ae_dev && ae_dev->ops && ae_dev->ops->flr_prepare)
+		ae_dev->ops->flr_prepare(ae_dev);
+}
+
+static void hns3_reset_done(struct pci_dev *pdev)
+{
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+
+	dev_info(&pdev->dev, "FLR done\n");
+	if (ae_dev && ae_dev->ops && ae_dev->ops->flr_done)
+		ae_dev->ops->flr_done(ae_dev);
+}
+
+static const struct pci_error_handlers hns3_err_handler = {
+	.error_detected = hns3_error_detected,
+	.slot_reset     = hns3_slot_reset,
+	.reset_prepare	= hns3_reset_prepare,
+	.reset_done	= hns3_reset_done,
+};
+
 static struct pci_driver hns3_driver = {
 	.name     = hns3_driver_name,
 	.id_table = hns3_pci_tbl,
 	.probe    = hns3_probe,
 	.remove   = hns3_remove,
+	.shutdown = hns3_shutdown,
 	.sriov_configure = hns3_pci_sriov_configure,
+	.err_handler    = &hns3_err_handler,
 };
 
 /* set default feature to hns3 */
@@ -1677,6 +2307,7 @@
 {
 	struct hnae3_handle *h = hns3_get_handle(netdev);
 	struct pci_dev *pdev = h->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
 
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 
@@ -1684,9 +2315,8 @@
 		NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
 		NETIF_F_GRO | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_GRE |
 		NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
-		NETIF_F_GSO_UDP_TUNNEL_CSUM;
-
-	netdev->hw_enc_features |= NETIF_F_TSO_MANGLEID;
+		NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC |
+		NETIF_F_TSO_MANGLEID | NETIF_F_FRAGLIST;
 
 	netdev->gso_partial_features |= NETIF_F_GSO_GRE_CSUM;
 
@@ -1696,30 +2326,47 @@
 		NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
 		NETIF_F_GRO | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_GRE |
 		NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
-		NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC |
+		NETIF_F_FRAGLIST;
 
 	netdev->vlan_features |=
 		NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM |
 		NETIF_F_SG | NETIF_F_GSO | NETIF_F_GRO |
 		NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_GRE |
 		NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
-		NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC |
+		NETIF_F_FRAGLIST;
 
 	netdev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
 		NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX |
 		NETIF_F_RXCSUM | NETIF_F_SG | NETIF_F_GSO |
 		NETIF_F_GRO | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_GRE |
 		NETIF_F_GSO_GRE_CSUM | NETIF_F_GSO_UDP_TUNNEL |
-		NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		NETIF_F_GSO_UDP_TUNNEL_CSUM | NETIF_F_SCTP_CRC |
+		NETIF_F_FRAGLIST;
 
-	if (pdev->revision != 0x20)
-		netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_FILTER;
+	if (ae_dev->dev_version >= HNAE3_DEVICE_VERSION_V2) {
+		netdev->hw_features |= NETIF_F_GRO_HW;
+		netdev->features |= NETIF_F_GRO_HW;
+
+		if (!(h->flags & HNAE3_SUPPORT_VF)) {
+			netdev->hw_features |= NETIF_F_NTUPLE;
+			netdev->features |= NETIF_F_NTUPLE;
+		}
+	}
+
+	if (test_bit(HNAE3_DEV_SUPPORT_UDP_GSO_B, ae_dev->caps)) {
+		netdev->hw_features |= NETIF_F_GSO_UDP_L4;
+		netdev->features |= NETIF_F_GSO_UDP_L4;
+		netdev->vlan_features |= NETIF_F_GSO_UDP_L4;
+		netdev->hw_enc_features |= NETIF_F_GSO_UDP_L4;
+	}
 }
 
 static int hns3_alloc_buffer(struct hns3_enet_ring *ring,
 			     struct hns3_desc_cb *cb)
 {
-	unsigned int order = hnae3_page_order(ring);
+	unsigned int order = hns3_page_order(ring);
 	struct page *p;
 
 	p = dev_alloc_pages(order);
@@ -1730,19 +2377,21 @@
 	cb->page_offset = 0;
 	cb->reuse_flag = 0;
 	cb->buf  = page_address(p);
-	cb->length = hnae3_page_size(ring);
+	cb->length = hns3_page_size(ring);
 	cb->type = DESC_TYPE_PAGE;
+	page_ref_add(p, USHRT_MAX - 1);
+	cb->pagecnt_bias = USHRT_MAX;
 
 	return 0;
 }
 
 static void hns3_free_buffer(struct hns3_enet_ring *ring,
-			     struct hns3_desc_cb *cb)
+			     struct hns3_desc_cb *cb, int budget)
 {
 	if (cb->type == DESC_TYPE_SKB)
-		dev_kfree_skb_any((struct sk_buff *)cb->priv);
-	else if (!HNAE3_IS_TX_RING(ring))
-		put_page((struct page *)cb->priv);
+		napi_consume_skb(cb->priv, budget);
+	else if (!HNAE3_IS_TX_RING(ring) && cb->pagecnt_bias)
+		__page_frag_cache_drain(cb->priv, cb->pagecnt_bias);
 	memset(cb, 0, sizeof(*cb));
 }
 
@@ -1751,7 +2400,7 @@
 	cb->dma = dma_map_page(ring_to_dev(ring), cb->priv, 0,
 			       cb->length, ring_to_dma_dir(ring));
 
-	if (dma_mapping_error(ring_to_dev(ring), cb->dma))
+	if (unlikely(dma_mapping_error(ring_to_dev(ring), cb->dma)))
 		return -EIO;
 
 	return 0;
@@ -1760,10 +2409,10 @@
 static void hns3_unmap_buffer(struct hns3_enet_ring *ring,
 			      struct hns3_desc_cb *cb)
 {
-	if (cb->type == DESC_TYPE_SKB)
+	if (cb->type == DESC_TYPE_SKB || cb->type == DESC_TYPE_FRAGLIST_SKB)
 		dma_unmap_single(ring_to_dev(ring), cb->dma, cb->length,
 				 ring_to_dma_dir(ring));
-	else
+	else if (cb->length)
 		dma_unmap_page(ring_to_dev(ring), cb->dma, cb->length,
 			       ring_to_dma_dir(ring));
 }
@@ -1772,9 +2421,11 @@
 {
 	hns3_unmap_buffer(ring, &ring->desc_cb[i]);
 	ring->desc[i].addr = 0;
+	ring->desc_cb[i].refill = 0;
 }
 
-static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i)
+static void hns3_free_buffer_detach(struct hns3_enet_ring *ring, int i,
+				    int budget)
 {
 	struct hns3_desc_cb *cb = &ring->desc_cb[i];
 
@@ -1782,7 +2433,7 @@
 		return;
 
 	hns3_buffer_detach(ring, i);
-	hns3_free_buffer(ring, cb);
+	hns3_free_buffer(ring, cb, budget);
 }
 
 static void hns3_free_buffers(struct hns3_enet_ring *ring)
@@ -1790,7 +2441,7 @@
 	int i;
 
 	for (i = 0; i < ring->desc_num; i++)
-		hns3_free_buffer_detach(ring, i);
+		hns3_free_buffer_detach(ring, i, 0);
 }
 
 /* free desc along with its attached buffer */
@@ -1811,16 +2462,15 @@
 {
 	int size = ring->desc_num * sizeof(ring->desc[0]);
 
-	ring->desc = dma_zalloc_coherent(ring_to_dev(ring), size,
-					 &ring->desc_dma_addr,
-					 GFP_KERNEL);
+	ring->desc = dma_alloc_coherent(ring_to_dev(ring), size,
+					&ring->desc_dma_addr, GFP_KERNEL);
 	if (!ring->desc)
 		return -ENOMEM;
 
 	return 0;
 }
 
-static int hns3_reserve_buffer_map(struct hns3_enet_ring *ring,
+static int hns3_alloc_and_map_buffer(struct hns3_enet_ring *ring,
 				   struct hns3_desc_cb *cb)
 {
 	int ret;
@@ -1836,19 +2486,20 @@
 	return 0;
 
 out_with_buf:
-	hns3_free_buffer(ring, cb);
+	hns3_free_buffer(ring, cb, 0);
 out:
 	return ret;
 }
 
-static int hns3_alloc_buffer_attach(struct hns3_enet_ring *ring, int i)
+static int hns3_alloc_and_attach_buffer(struct hns3_enet_ring *ring, int i)
 {
-	int ret = hns3_reserve_buffer_map(ring, &ring->desc_cb[i]);
+	int ret = hns3_alloc_and_map_buffer(ring, &ring->desc_cb[i]);
 
 	if (ret)
 		return ret;
 
 	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma);
+	ring->desc_cb[i].refill = 1;
 
 	return 0;
 }
@@ -1859,7 +2510,7 @@
 	int i, j, ret;
 
 	for (i = 0; i < ring->desc_num; i++) {
-		ret = hns3_alloc_buffer_attach(ring, i);
+		ret = hns3_alloc_and_attach_buffer(ring, i);
 		if (ret)
 			goto out_buffer_fail;
 	}
@@ -1868,84 +2519,91 @@
 
 out_buffer_fail:
 	for (j = i - 1; j >= 0; j--)
-		hns3_free_buffer_detach(ring, j);
+		hns3_free_buffer_detach(ring, j, 0);
 	return ret;
 }
 
-/* detach a in-used buffer and replace with a reserved one  */
+/* detach a in-used buffer and replace with a reserved one */
 static void hns3_replace_buffer(struct hns3_enet_ring *ring, int i,
 				struct hns3_desc_cb *res_cb)
 {
 	hns3_unmap_buffer(ring, &ring->desc_cb[i]);
 	ring->desc_cb[i] = *res_cb;
 	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma);
+	ring->desc_cb[i].refill = 1;
 	ring->desc[i].rx.bd_base_info = 0;
 }
 
 static void hns3_reuse_buffer(struct hns3_enet_ring *ring, int i)
 {
 	ring->desc_cb[i].reuse_flag = 0;
-	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma
-		+ ring->desc_cb[i].page_offset);
+	ring->desc_cb[i].refill = 1;
+	ring->desc[i].addr = cpu_to_le64(ring->desc_cb[i].dma +
+					 ring->desc_cb[i].page_offset);
 	ring->desc[i].rx.bd_base_info = 0;
+
+	dma_sync_single_for_device(ring_to_dev(ring),
+			ring->desc_cb[i].dma + ring->desc_cb[i].page_offset,
+			hns3_buf_size(ring),
+			DMA_FROM_DEVICE);
 }
 
-static void hns3_nic_reclaim_one_desc(struct hns3_enet_ring *ring, int *bytes,
-				      int *pkts)
+static bool hns3_nic_reclaim_desc(struct hns3_enet_ring *ring,
+				  int *bytes, int *pkts, int budget)
 {
-	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_clean];
+	/* pair with ring->last_to_use update in hns3_tx_doorbell(),
+	 * smp_store_release() is not used in hns3_tx_doorbell() because
+	 * the doorbell operation already have the needed barrier operation.
+	 */
+	int ltu = smp_load_acquire(&ring->last_to_use);
+	int ntc = ring->next_to_clean;
+	struct hns3_desc_cb *desc_cb;
+	bool reclaimed = false;
+	struct hns3_desc *desc;
 
-	(*pkts) += (desc_cb->type == DESC_TYPE_SKB);
-	(*bytes) += desc_cb->length;
-	/* desc_cb will be cleaned, after hnae3_free_buffer_detach*/
-	hns3_free_buffer_detach(ring, ring->next_to_clean);
+	while (ltu != ntc) {
+		desc = &ring->desc[ntc];
 
-	ring_ptr_move_fw(ring, next_to_clean);
+		if (le16_to_cpu(desc->tx.bdtp_fe_sc_vld_ra_ri) &
+				BIT(HNS3_TXD_VLD_B))
+			break;
+
+		desc_cb = &ring->desc_cb[ntc];
+		(*pkts) += (desc_cb->type == DESC_TYPE_SKB);
+		(*bytes) += desc_cb->length;
+		/* desc_cb will be cleaned, after hnae3_free_buffer_detach */
+		hns3_free_buffer_detach(ring, ntc, budget);
+
+		if (++ntc == ring->desc_num)
+			ntc = 0;
+
+		/* Issue prefetch for next Tx descriptor */
+		prefetch(&ring->desc_cb[ntc]);
+		reclaimed = true;
+	}
+
+	if (unlikely(!reclaimed))
+		return false;
+
+	/* This smp_store_release() pairs with smp_load_acquire() in
+	 * ring_space called by hns3_nic_net_xmit.
+	 */
+	smp_store_release(&ring->next_to_clean, ntc);
+	return true;
 }
 
-static int is_valid_clean_head(struct hns3_enet_ring *ring, int h)
+void hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget)
 {
-	int u = ring->next_to_use;
-	int c = ring->next_to_clean;
-
-	if (unlikely(h > ring->desc_num))
-		return 0;
-
-	return u > c ? (h > c && h <= u) : (h > c || h <= u);
-}
-
-bool hns3_clean_tx_ring(struct hns3_enet_ring *ring, int budget)
-{
-	struct net_device *netdev = ring->tqp->handle->kinfo.netdev;
+	struct net_device *netdev = ring_to_netdev(ring);
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	struct netdev_queue *dev_queue;
 	int bytes, pkts;
-	int head;
-
-	head = readl_relaxed(ring->tqp->io_base + HNS3_RING_TX_RING_HEAD_REG);
-	rmb(); /* Make sure head is ready before touch any data */
-
-	if (is_ring_empty(ring) || head == ring->next_to_clean)
-		return true; /* no data to poll */
-
-	if (unlikely(!is_valid_clean_head(ring, head))) {
-		netdev_err(netdev, "wrong head (%d, %d-%d)\n", head,
-			   ring->next_to_use, ring->next_to_clean);
-
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.io_err_cnt++;
-		u64_stats_update_end(&ring->syncp);
-		return true;
-	}
 
 	bytes = 0;
 	pkts = 0;
-	while (head != ring->next_to_clean && budget) {
-		hns3_nic_reclaim_one_desc(ring, &bytes, &pkts);
-		/* Issue prefetch for next Tx descriptor */
-		prefetch(&ring->desc_cb[ring->next_to_clean]);
-		budget--;
-	}
+
+	if (unlikely(!hns3_nic_reclaim_desc(ring, &bytes, &pkts, budget)))
+		return;
 
 	ring->tqp_vector->tx_group.total_bytes += bytes;
 	ring->tqp_vector->tx_group.total_packets += pkts;
@@ -1958,8 +2616,8 @@
 	dev_queue = netdev_get_tx_queue(netdev, ring->tqp->tqp_index);
 	netdev_tx_completed_queue(dev_queue, pkts, bytes);
 
-	if (unlikely(pkts && netif_carrier_ok(netdev) &&
-		     (ring_space(ring) > HNS3_MAX_BD_PER_PKT))) {
+	if (unlikely(netif_carrier_ok(netdev) &&
+		     ring_space(ring) > HNS3_MAX_TSO_BD_NUM)) {
 		/* Make sure that anybody stopping the queue after this
 		 * sees the new next_to_clean.
 		 */
@@ -1970,8 +2628,6 @@
 			ring->stats.restart_queue++;
 		}
 	}
-
-	return !!budget;
 }
 
 static int hns3_desc_unused(struct hns3_enet_ring *ring)
@@ -1979,11 +2635,15 @@
 	int ntc = ring->next_to_clean;
 	int ntu = ring->next_to_use;
 
+	if (unlikely(ntc == ntu && !ring->desc_cb[ntc].refill))
+		return ring->desc_num;
+
 	return ((ntc >= ntu) ? 0 : ring->desc_num) + ntc - ntu;
 }
 
-static void
-hns3_nic_alloc_rx_buffers(struct hns3_enet_ring *ring, int cleand_count)
+/* Return true if there is any allocation failure */
+static bool hns3_nic_alloc_rx_buffers(struct hns3_enet_ring *ring,
+				      int cleand_count)
 {
 	struct hns3_desc_cb *desc_cb;
 	struct hns3_desc_cb res_cbs;
@@ -1998,88 +2658,149 @@
 
 			hns3_reuse_buffer(ring, ring->next_to_use);
 		} else {
-			ret = hns3_reserve_buffer_map(ring, &res_cbs);
+			ret = hns3_alloc_and_map_buffer(ring, &res_cbs);
 			if (ret) {
 				u64_stats_update_begin(&ring->syncp);
 				ring->stats.sw_err_cnt++;
 				u64_stats_update_end(&ring->syncp);
 
-				netdev_err(ring->tqp->handle->kinfo.netdev,
-					   "hnae reserve buffer map failed.\n");
-				break;
+				hns3_rl_err(ring_to_netdev(ring),
+					    "alloc rx buffer failed: %d\n",
+					    ret);
+
+				writel(i, ring->tqp->io_base +
+				       HNS3_RING_RX_RING_HEAD_REG);
+				return true;
 			}
 			hns3_replace_buffer(ring, ring->next_to_use, &res_cbs);
+
+			u64_stats_update_begin(&ring->syncp);
+			ring->stats.non_reuse_pg++;
+			u64_stats_update_end(&ring->syncp);
 		}
 
 		ring_ptr_move_fw(ring, next_to_use);
 	}
 
-	wmb(); /* Make all data has been write before submit */
-	writel_relaxed(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG);
+	writel(i, ring->tqp->io_base + HNS3_RING_RX_RING_HEAD_REG);
+	return false;
+}
+
+static bool hns3_page_is_reusable(struct page *page)
+{
+	return page_to_nid(page) == numa_mem_id() &&
+		!page_is_pfmemalloc(page);
+}
+
+static bool hns3_can_reuse_page(struct hns3_desc_cb *cb)
+{
+	return (page_count(cb->priv) - cb->pagecnt_bias) == 1;
 }
 
 static void hns3_nic_reuse_page(struct sk_buff *skb, int i,
 				struct hns3_enet_ring *ring, int pull_len,
 				struct hns3_desc_cb *desc_cb)
 {
-	struct hns3_desc *desc;
-	u32 truesize;
-	int size;
-	int last_offset;
-	bool twobufs;
+	struct hns3_desc *desc = &ring->desc[ring->next_to_clean];
+	int size = le16_to_cpu(desc->rx.size);
+	u32 truesize = hns3_buf_size(ring);
 
-	twobufs = ((PAGE_SIZE < 8192) &&
-		hnae3_buf_size(ring) == HNS3_BUFFER_SIZE_2048);
-
-	desc = &ring->desc[ring->next_to_clean];
-	size = le16_to_cpu(desc->rx.size);
-
-	truesize = hnae3_buf_size(ring);
-
-	if (!twobufs)
-		last_offset = hnae3_page_size(ring) - hnae3_buf_size(ring);
-
+	desc_cb->pagecnt_bias--;
 	skb_add_rx_frag(skb, i, desc_cb->priv, desc_cb->page_offset + pull_len,
 			size - pull_len, truesize);
 
-	 /* Avoid re-using remote pages,flag default unreuse */
-	if (unlikely(page_to_nid(desc_cb->priv) != numa_node_id()))
-		return;
-
-	if (twobufs) {
-		/* If we are only owner of page we can reuse it */
-		if (likely(page_count(desc_cb->priv) == 1)) {
-			/* Flip page offset to other buffer */
-			desc_cb->page_offset ^= truesize;
-
-			desc_cb->reuse_flag = 1;
-			/* bump ref count on page before it is given*/
-			get_page(desc_cb->priv);
-		}
+	/* Avoid re-using remote pages, or the stack is still using the page
+	 * when page_offset rollback to zero, flag default unreuse
+	 */
+	if (unlikely(!hns3_page_is_reusable(desc_cb->priv)) ||
+	    (!desc_cb->page_offset && !hns3_can_reuse_page(desc_cb))) {
+		__page_frag_cache_drain(desc_cb->priv, desc_cb->pagecnt_bias);
 		return;
 	}
 
 	/* Move offset up to the next cache line */
 	desc_cb->page_offset += truesize;
 
-	if (desc_cb->page_offset <= last_offset) {
+	if (desc_cb->page_offset + truesize <= hns3_page_size(ring)) {
 		desc_cb->reuse_flag = 1;
-		/* Bump ref count on page before it is given*/
-		get_page(desc_cb->priv);
+	} else if (hns3_can_reuse_page(desc_cb)) {
+		desc_cb->reuse_flag = 1;
+		desc_cb->page_offset = 0;
+	} else if (desc_cb->pagecnt_bias) {
+		__page_frag_cache_drain(desc_cb->priv, desc_cb->pagecnt_bias);
+		return;
+	}
+
+	if (unlikely(!desc_cb->pagecnt_bias)) {
+		page_ref_add(desc_cb->priv, USHRT_MAX);
+		desc_cb->pagecnt_bias = USHRT_MAX;
 	}
 }
 
-static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb,
-			     struct hns3_desc *desc)
+static int hns3_gro_complete(struct sk_buff *skb, u32 l234info)
 {
-	struct net_device *netdev = ring->tqp->handle->kinfo.netdev;
-	int l3_type, l4_type;
-	u32 bd_base_info;
-	int ol4_type;
-	u32 l234info;
+	__be16 type = skb->protocol;
+	struct tcphdr *th;
+	int depth = 0;
 
-	bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
-	l234info = le32_to_cpu(desc->rx.l234_info);
+	while (eth_type_vlan(type)) {
+		struct vlan_hdr *vh;
+
+		if ((depth + VLAN_HLEN) > skb_headlen(skb))
+			return -EFAULT;
+
+		vh = (struct vlan_hdr *)(skb->data + depth);
+		type = vh->h_vlan_encapsulated_proto;
+		depth += VLAN_HLEN;
+	}
+
+	skb_set_network_header(skb, depth);
+
+	if (type == htons(ETH_P_IP)) {
+		const struct iphdr *iph = ip_hdr(skb);
+
+		depth += sizeof(struct iphdr);
+		skb_set_transport_header(skb, depth);
+		th = tcp_hdr(skb);
+		th->check = ~tcp_v4_check(skb->len - depth, iph->saddr,
+					  iph->daddr, 0);
+	} else if (type == htons(ETH_P_IPV6)) {
+		const struct ipv6hdr *iph = ipv6_hdr(skb);
+
+		depth += sizeof(struct ipv6hdr);
+		skb_set_transport_header(skb, depth);
+		th = tcp_hdr(skb);
+		th->check = ~tcp_v6_check(skb->len - depth, &iph->saddr,
+					  &iph->daddr, 0);
+	} else {
+		hns3_rl_err(skb->dev,
+			    "Error: FW GRO supports only IPv4/IPv6, not 0x%04x, depth: %d\n",
+			    be16_to_cpu(type), depth);
+		return -EFAULT;
+	}
+
+	skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+	if (th->cwr)
+		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ECN;
+
+	if (l234info & BIT(HNS3_RXD_GRO_FIXID_B))
+		skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_FIXEDID;
+
+	skb->csum_start = (unsigned char *)th - skb->head;
+	skb->csum_offset = offsetof(struct tcphdr, check);
+	skb->ip_summed = CHECKSUM_PARTIAL;
+
+	trace_hns3_gro(skb);
+
+	return 0;
+}
+
+static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb,
+			     u32 l234info, u32 bd_base_info, u32 ol_info)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	int l3_type, l4_type;
+	int ol4_type;
 
 	skb->ip_summed = CHECKSUM_NONE;
 
@@ -2089,14 +2810,12 @@
 		return;
 
 	/* check if hardware has done checksum */
-	if (!hnae3_get_bit(bd_base_info, HNS3_RXD_L3L4P_B))
+	if (!(bd_base_info & BIT(HNS3_RXD_L3L4P_B)))
 		return;
 
-	if (unlikely(hnae3_get_bit(l234info, HNS3_RXD_L3E_B) ||
-		     hnae3_get_bit(l234info, HNS3_RXD_L4E_B) ||
-		     hnae3_get_bit(l234info, HNS3_RXD_OL3E_B) ||
-		     hnae3_get_bit(l234info, HNS3_RXD_OL4E_B))) {
-		netdev_err(netdev, "L3/L4 error pkt\n");
+	if (unlikely(l234info & (BIT(HNS3_RXD_L3E_B) | BIT(HNS3_RXD_L4E_B) |
+				 BIT(HNS3_RXD_OL3E_B) |
+				 BIT(HNS3_RXD_OL4E_B)))) {
 		u64_stats_update_begin(&ring->syncp);
 		ring->stats.l3l4_csum_err++;
 		u64_stats_update_end(&ring->syncp);
@@ -2104,19 +2823,19 @@
 		return;
 	}
 
-	l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M,
-				  HNS3_RXD_L3ID_S);
-	l4_type = hnae3_get_field(l234info, HNS3_RXD_L4ID_M,
-				  HNS3_RXD_L4ID_S);
-
-	ol4_type = hnae3_get_field(l234info, HNS3_RXD_OL4ID_M,
+	ol4_type = hnae3_get_field(ol_info, HNS3_RXD_OL4ID_M,
 				   HNS3_RXD_OL4ID_S);
 	switch (ol4_type) {
 	case HNS3_OL4_TYPE_MAC_IN_UDP:
 	case HNS3_OL4_TYPE_NVGRE:
 		skb->csum_level = 1;
-		/* fall through */
+		fallthrough;
 	case HNS3_OL4_TYPE_NO_TUN:
+		l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M,
+					  HNS3_RXD_L3ID_S);
+		l4_type = hnae3_get_field(l234info, HNS3_RXD_L4ID_M,
+					  HNS3_RXD_L4ID_S);
+
 		/* Can checksum ipv4 or ipv6 + UDP/TCP/SCTP packets */
 		if ((l3_type == HNS3_L3_TYPE_IPV4 ||
 		     l3_type == HNS3_L3_TYPE_IPV6) &&
@@ -2125,11 +2844,16 @@
 		     l4_type == HNS3_L4_TYPE_SCTP))
 			skb->ip_summed = CHECKSUM_UNNECESSARY;
 		break;
+	default:
+		break;
 	}
 }
 
 static void hns3_rx_skb(struct hns3_enet_ring *ring, struct sk_buff *skb)
 {
+	if (skb_has_frag_list(skb))
+		napi_gro_flush(&ring->tqp_vector->napi, false);
+
 	napi_gro_receive(&ring->tqp_vector->napi, skb);
 }
 
@@ -2137,9 +2861,11 @@
 				struct hns3_desc *desc, u32 l234info,
 				u16 *vlan_tag)
 {
+	struct hnae3_handle *handle = ring->tqp->handle;
 	struct pci_dev *pdev = ring->tqp->handle->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
 
-	if (pdev->revision == 0x20) {
+	if (unlikely(ae_dev->dev_version < HNAE3_DEVICE_VERSION_V2)) {
 		*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
 		if (!(*vlan_tag & VLAN_VID_MASK))
 			*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
@@ -2149,64 +2875,63 @@
 
 #define HNS3_STRP_OUTER_VLAN	0x1
 #define HNS3_STRP_INNER_VLAN	0x2
+#define HNS3_STRP_BOTH		0x3
 
+	/* Hardware always insert VLAN tag into RX descriptor when
+	 * remove the tag from packet, driver needs to determine
+	 * reporting which tag to stack.
+	 */
 	switch (hnae3_get_field(l234info, HNS3_RXD_STRP_TAGP_M,
 				HNS3_RXD_STRP_TAGP_S)) {
 	case HNS3_STRP_OUTER_VLAN:
+		if (handle->port_base_vlan_state !=
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			return false;
+
 		*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
 		return true;
 	case HNS3_STRP_INNER_VLAN:
+		if (handle->port_base_vlan_state !=
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			return false;
+
 		*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
+		return true;
+	case HNS3_STRP_BOTH:
+		if (handle->port_base_vlan_state ==
+				HNAE3_PORT_BASE_VLAN_DISABLE)
+			*vlan_tag = le16_to_cpu(desc->rx.ot_vlan_tag);
+		else
+			*vlan_tag = le16_to_cpu(desc->rx.vlan_tag);
+
 		return true;
 	default:
 		return false;
 	}
 }
 
-static int hns3_handle_rx_bd(struct hns3_enet_ring *ring,
-			     struct sk_buff **out_skb, int *out_bnum)
+static void hns3_rx_ring_move_fw(struct hns3_enet_ring *ring)
 {
-	struct net_device *netdev = ring->tqp->handle->kinfo.netdev;
-	struct hns3_desc_cb *desc_cb;
-	struct hns3_desc *desc;
+	ring->desc[ring->next_to_clean].rx.bd_base_info &=
+		cpu_to_le32(~BIT(HNS3_RXD_VLD_B));
+	ring->desc_cb[ring->next_to_clean].refill = 0;
+	ring->next_to_clean += 1;
+
+	if (unlikely(ring->next_to_clean == ring->desc_num))
+		ring->next_to_clean = 0;
+}
+
+static int hns3_alloc_skb(struct hns3_enet_ring *ring, unsigned int length,
+			  unsigned char *va)
+{
+	struct hns3_desc_cb *desc_cb = &ring->desc_cb[ring->next_to_clean];
+	struct net_device *netdev = ring_to_netdev(ring);
 	struct sk_buff *skb;
-	unsigned char *va;
-	u32 bd_base_info;
-	int pull_len;
-	u32 l234info;
-	int length;
-	int bnum;
 
-	desc = &ring->desc[ring->next_to_clean];
-	desc_cb = &ring->desc_cb[ring->next_to_clean];
-
-	prefetch(desc);
-
-	length = le16_to_cpu(desc->rx.size);
-	bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
-
-	/* Check valid BD */
-	if (unlikely(!hnae3_get_bit(bd_base_info, HNS3_RXD_VLD_B)))
-		return -EFAULT;
-
-	va = (unsigned char *)desc_cb->buf + desc_cb->page_offset;
-
-	/* Prefetch first cache line of first page
-	 * Idea is to cache few bytes of the header of the packet. Our L1 Cache
-	 * line size is 64B so need to prefetch twice to make it 128B. But in
-	 * actual we can have greater size of caches with 128B Level 1 cache
-	 * lines. In such a case, single fetch would suffice to cache in the
-	 * relevant part of the header.
-	 */
-	prefetch(va);
-#if L1_CACHE_BYTES < 128
-	prefetch(va + L1_CACHE_BYTES);
-#endif
-
-	skb = *out_skb = napi_alloc_skb(&ring->tqp_vector->napi,
-					HNS3_RX_HEAD_SIZE);
+	ring->skb = napi_alloc_skb(&ring->tqp_vector->napi, HNS3_RX_HEAD_SIZE);
+	skb = ring->skb;
 	if (unlikely(!skb)) {
-		netdev_err(netdev, "alloc rx skb fail\n");
+		hns3_rl_err(netdev, "alloc rx skb fail\n");
 
 		u64_stats_update_begin(&ring->syncp);
 		ring->stats.sw_err_cnt++;
@@ -2215,45 +2940,158 @@
 		return -ENOMEM;
 	}
 
+	trace_hns3_rx_desc(ring);
 	prefetchw(skb->data);
 
-	bnum = 1;
+	ring->pending_buf = 1;
+	ring->frag_num = 0;
+	ring->tail_skb = NULL;
 	if (length <= HNS3_RX_HEAD_SIZE) {
 		memcpy(__skb_put(skb, length), va, ALIGN(length, sizeof(long)));
 
 		/* We can reuse buffer as-is, just make sure it is local */
-		if (likely(page_to_nid(desc_cb->priv) == numa_node_id()))
+		if (likely(hns3_page_is_reusable(desc_cb->priv)))
 			desc_cb->reuse_flag = 1;
 		else /* This page cannot be reused so discard it */
-			put_page(desc_cb->priv);
+			__page_frag_cache_drain(desc_cb->priv,
+						desc_cb->pagecnt_bias);
 
-		ring_ptr_move_fw(ring, next_to_clean);
-	} else {
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.seg_pkt_cnt++;
-		u64_stats_update_end(&ring->syncp);
+		hns3_rx_ring_move_fw(ring);
+		return 0;
+	}
+	u64_stats_update_begin(&ring->syncp);
+	ring->stats.seg_pkt_cnt++;
+	u64_stats_update_end(&ring->syncp);
 
-		pull_len = eth_get_headlen(va, HNS3_RX_HEAD_SIZE);
+	ring->pull_len = eth_get_headlen(netdev, va, HNS3_RX_HEAD_SIZE);
+	__skb_put(skb, ring->pull_len);
+	hns3_nic_reuse_page(skb, ring->frag_num++, ring, ring->pull_len,
+			    desc_cb);
+	hns3_rx_ring_move_fw(ring);
 
-		memcpy(__skb_put(skb, pull_len), va,
-		       ALIGN(pull_len, sizeof(long)));
+	return 0;
+}
 
-		hns3_nic_reuse_page(skb, 0, ring, pull_len, desc_cb);
-		ring_ptr_move_fw(ring, next_to_clean);
+static int hns3_add_frag(struct hns3_enet_ring *ring)
+{
+	struct sk_buff *skb = ring->skb;
+	struct sk_buff *head_skb = skb;
+	struct sk_buff *new_skb;
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_desc *desc;
+	u32 bd_base_info;
 
-		while (!hnae3_get_bit(bd_base_info, HNS3_RXD_FE_B)) {
-			desc = &ring->desc[ring->next_to_clean];
-			desc_cb = &ring->desc_cb[ring->next_to_clean];
-			bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
-			hns3_nic_reuse_page(skb, bnum, ring, 0, desc_cb);
-			ring_ptr_move_fw(ring, next_to_clean);
-			bnum++;
+	do {
+		desc = &ring->desc[ring->next_to_clean];
+		desc_cb = &ring->desc_cb[ring->next_to_clean];
+		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
+		/* make sure HW write desc complete */
+		dma_rmb();
+		if (!(bd_base_info & BIT(HNS3_RXD_VLD_B)))
+			return -ENXIO;
+
+		if (unlikely(ring->frag_num >= MAX_SKB_FRAGS)) {
+			new_skb = napi_alloc_skb(&ring->tqp_vector->napi, 0);
+			if (unlikely(!new_skb)) {
+				hns3_rl_err(ring_to_netdev(ring),
+					    "alloc rx fraglist skb fail\n");
+				return -ENXIO;
+			}
+			ring->frag_num = 0;
+
+			if (ring->tail_skb) {
+				ring->tail_skb->next = new_skb;
+				ring->tail_skb = new_skb;
+			} else {
+				skb_shinfo(skb)->frag_list = new_skb;
+				ring->tail_skb = new_skb;
+			}
 		}
+
+		if (ring->tail_skb) {
+			head_skb->truesize += hns3_buf_size(ring);
+			head_skb->data_len += le16_to_cpu(desc->rx.size);
+			head_skb->len += le16_to_cpu(desc->rx.size);
+			skb = ring->tail_skb;
+		}
+
+		dma_sync_single_for_cpu(ring_to_dev(ring),
+				desc_cb->dma + desc_cb->page_offset,
+				hns3_buf_size(ring),
+				DMA_FROM_DEVICE);
+
+		hns3_nic_reuse_page(skb, ring->frag_num++, ring, 0, desc_cb);
+		trace_hns3_rx_desc(ring);
+		hns3_rx_ring_move_fw(ring);
+		ring->pending_buf++;
+	} while (!(bd_base_info & BIT(HNS3_RXD_FE_B)));
+
+	return 0;
+}
+
+static int hns3_set_gro_and_checksum(struct hns3_enet_ring *ring,
+				     struct sk_buff *skb, u32 l234info,
+				     u32 bd_base_info, u32 ol_info)
+{
+	u32 l3_type;
+
+	skb_shinfo(skb)->gso_size = hnae3_get_field(bd_base_info,
+						    HNS3_RXD_GRO_SIZE_M,
+						    HNS3_RXD_GRO_SIZE_S);
+	/* if there is no HW GRO, do not set gro params */
+	if (!skb_shinfo(skb)->gso_size) {
+		hns3_rx_checksum(ring, skb, l234info, bd_base_info, ol_info);
+		return 0;
 	}
 
-	*out_bnum = bnum;
+	NAPI_GRO_CB(skb)->count = hnae3_get_field(l234info,
+						  HNS3_RXD_GRO_COUNT_M,
+						  HNS3_RXD_GRO_COUNT_S);
 
+	l3_type = hnae3_get_field(l234info, HNS3_RXD_L3ID_M, HNS3_RXD_L3ID_S);
+	if (l3_type == HNS3_L3_TYPE_IPV4)
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+	else if (l3_type == HNS3_L3_TYPE_IPV6)
+		skb_shinfo(skb)->gso_type = SKB_GSO_TCPV6;
+	else
+		return -EFAULT;
+
+	return  hns3_gro_complete(skb, l234info);
+}
+
+static void hns3_set_rx_skb_rss_type(struct hns3_enet_ring *ring,
+				     struct sk_buff *skb, u32 rss_hash)
+{
+	struct hnae3_handle *handle = ring->tqp->handle;
+	enum pkt_hash_types rss_type;
+
+	if (rss_hash)
+		rss_type = handle->kinfo.rss_type;
+	else
+		rss_type = PKT_HASH_TYPE_NONE;
+
+	skb_set_hash(skb, rss_hash, rss_type);
+}
+
+static int hns3_handle_bdinfo(struct hns3_enet_ring *ring, struct sk_buff *skb)
+{
+	struct net_device *netdev = ring_to_netdev(ring);
+	enum hns3_pkt_l2t_type l2_frame_type;
+	u32 bd_base_info, l234info, ol_info;
+	struct hns3_desc *desc;
+	unsigned int len;
+	int pre_ntc, ret;
+
+	/* bdinfo handled below is only valid on the last BD of the
+	 * current packet, and ring->next_to_clean indicates the first
+	 * descriptor of next packet, so need - 1 below.
+	 */
+	pre_ntc = ring->next_to_clean ? (ring->next_to_clean - 1) :
+					(ring->desc_num - 1);
+	desc = &ring->desc[pre_ntc];
+	bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
 	l234info = le32_to_cpu(desc->rx.l234_info);
+	ol_info = le32_to_cpu(desc->rx.ol_info);
 
 	/* Based on hw strategy, the tag offloaded will be stored at
 	 * ot_vlan_tag in two layer tag case, and stored at vlan_tag
@@ -2263,138 +3101,180 @@
 		u16 vlan_tag;
 
 		if (hns3_parse_vlan_tag(ring, desc, l234info, &vlan_tag))
-			__vlan_hwaccel_put_tag(skb,
-					       htons(ETH_P_8021Q),
+			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
 					       vlan_tag);
 	}
 
-	if (unlikely(!hnae3_get_bit(bd_base_info, HNS3_RXD_VLD_B))) {
-		netdev_err(netdev, "no valid bd,%016llx,%016llx\n",
-			   ((u64 *)desc)[0], ((u64 *)desc)[1]);
+	if (unlikely(!desc->rx.pkt_len || (l234info & (BIT(HNS3_RXD_TRUNCAT_B) |
+				  BIT(HNS3_RXD_L2E_B))))) {
 		u64_stats_update_begin(&ring->syncp);
-		ring->stats.non_vld_descs++;
+		if (l234info & BIT(HNS3_RXD_L2E_B))
+			ring->stats.l2_err++;
+		else
+			ring->stats.err_pkt_len++;
 		u64_stats_update_end(&ring->syncp);
 
-		dev_kfree_skb_any(skb);
-		return -EINVAL;
-	}
-
-	if (unlikely((!desc->rx.pkt_len) ||
-		     hnae3_get_bit(l234info, HNS3_RXD_TRUNCAT_B))) {
-		netdev_err(netdev, "truncated pkt\n");
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.err_pkt_len++;
-		u64_stats_update_end(&ring->syncp);
-
-		dev_kfree_skb_any(skb);
 		return -EFAULT;
 	}
 
-	if (unlikely(hnae3_get_bit(l234info, HNS3_RXD_L2E_B))) {
-		netdev_err(netdev, "L2 error pkt\n");
-		u64_stats_update_begin(&ring->syncp);
-		ring->stats.l2_err++;
-		u64_stats_update_end(&ring->syncp);
+	len = skb->len;
 
-		dev_kfree_skb_any(skb);
-		return -EFAULT;
+	/* Do update ip stack process */
+	skb->protocol = eth_type_trans(skb, netdev);
+
+	/* This is needed in order to enable forwarding support */
+	ret = hns3_set_gro_and_checksum(ring, skb, l234info,
+					bd_base_info, ol_info);
+	if (unlikely(ret)) {
+		u64_stats_update_begin(&ring->syncp);
+		ring->stats.rx_err_cnt++;
+		u64_stats_update_end(&ring->syncp);
+		return ret;
 	}
+
+	l2_frame_type = hnae3_get_field(l234info, HNS3_RXD_DMAC_M,
+					HNS3_RXD_DMAC_S);
 
 	u64_stats_update_begin(&ring->syncp);
 	ring->stats.rx_pkts++;
-	ring->stats.rx_bytes += skb->len;
+	ring->stats.rx_bytes += len;
+
+	if (l2_frame_type == HNS3_L2_TYPE_MULTICAST)
+		ring->stats.rx_multicast++;
+
 	u64_stats_update_end(&ring->syncp);
 
-	ring->tqp_vector->rx_group.total_bytes += skb->len;
+	ring->tqp_vector->rx_group.total_bytes += len;
 
-	hns3_rx_checksum(ring, skb, desc);
+	hns3_set_rx_skb_rss_type(ring, skb, le32_to_cpu(desc->rx.rss_hash));
 	return 0;
 }
 
-int hns3_clean_rx_ring(
-		struct hns3_enet_ring *ring, int budget,
-		void (*rx_fn)(struct hns3_enet_ring *, struct sk_buff *))
+static int hns3_handle_rx_bd(struct hns3_enet_ring *ring)
+{
+	struct sk_buff *skb = ring->skb;
+	struct hns3_desc_cb *desc_cb;
+	struct hns3_desc *desc;
+	unsigned int length;
+	u32 bd_base_info;
+	int ret;
+
+	desc = &ring->desc[ring->next_to_clean];
+	desc_cb = &ring->desc_cb[ring->next_to_clean];
+
+	prefetch(desc);
+
+	if (!skb) {
+		bd_base_info = le32_to_cpu(desc->rx.bd_base_info);
+
+		/* Check valid BD */
+		if (unlikely(!(bd_base_info & BIT(HNS3_RXD_VLD_B))))
+			return -ENXIO;
+
+		dma_rmb();
+		length = le16_to_cpu(desc->rx.size);
+
+		ring->va = desc_cb->buf + desc_cb->page_offset;
+
+		dma_sync_single_for_cpu(ring_to_dev(ring),
+				desc_cb->dma + desc_cb->page_offset,
+				hns3_buf_size(ring),
+				DMA_FROM_DEVICE);
+
+		/* Prefetch first cache line of first page.
+		 * Idea is to cache few bytes of the header of the packet.
+		 * Our L1 Cache line size is 64B so need to prefetch twice to make
+		 * it 128B. But in actual we can have greater size of caches with
+		 * 128B Level 1 cache lines. In such a case, single fetch would
+		 * suffice to cache in the relevant part of the header.
+		 */
+		net_prefetch(ring->va);
+
+		ret = hns3_alloc_skb(ring, length, ring->va);
+		skb = ring->skb;
+
+		if (ret < 0) /* alloc buffer fail */
+			return ret;
+		if (!(bd_base_info & BIT(HNS3_RXD_FE_B))) { /* need add frag */
+			ret = hns3_add_frag(ring);
+			if (ret)
+				return ret;
+		}
+	} else {
+		ret = hns3_add_frag(ring);
+		if (ret)
+			return ret;
+	}
+
+	/* As the head data may be changed when GRO enable, copy
+	 * the head data in after other data rx completed
+	 */
+	if (skb->len > HNS3_RX_HEAD_SIZE)
+		memcpy(skb->data, ring->va,
+		       ALIGN(ring->pull_len, sizeof(long)));
+
+	ret = hns3_handle_bdinfo(ring, skb);
+	if (unlikely(ret)) {
+		dev_kfree_skb_any(skb);
+		return ret;
+	}
+
+	skb_record_rx_queue(skb, ring->tqp->tqp_index);
+	return 0;
+}
+
+int hns3_clean_rx_ring(struct hns3_enet_ring *ring, int budget,
+		       void (*rx_fn)(struct hns3_enet_ring *, struct sk_buff *))
 {
 #define RCB_NOF_ALLOC_RX_BUFF_ONCE 16
-	struct net_device *netdev = ring->tqp->handle->kinfo.netdev;
-	int recv_pkts, recv_bds, clean_count, err;
 	int unused_count = hns3_desc_unused(ring);
-	struct sk_buff *skb = NULL;
-	int num, bnum = 0;
+	bool failure = false;
+	int recv_pkts = 0;
+	int err;
 
-	num = readl_relaxed(ring->tqp->io_base + HNS3_RING_RX_RING_FBDNUM_REG);
-	rmb(); /* Make sure num taken effect before the other data is touched */
+	unused_count -= ring->pending_buf;
 
-	recv_pkts = 0, recv_bds = 0, clean_count = 0;
-	num -= unused_count;
-
-	while (recv_pkts < budget && recv_bds < num) {
+	while (recv_pkts < budget) {
 		/* Reuse or realloc buffers */
-		if (clean_count + unused_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) {
-			hns3_nic_alloc_rx_buffers(ring,
-						  clean_count + unused_count);
-			clean_count = 0;
-			unused_count = hns3_desc_unused(ring);
+		if (unused_count >= RCB_NOF_ALLOC_RX_BUFF_ONCE) {
+			failure = failure ||
+				hns3_nic_alloc_rx_buffers(ring, unused_count);
+			unused_count = 0;
 		}
 
 		/* Poll one pkt */
-		err = hns3_handle_rx_bd(ring, &skb, &bnum);
-		if (unlikely(!skb)) /* This fault cannot be repaired */
+		err = hns3_handle_rx_bd(ring);
+		/* Do not get FE for the packet or failed to alloc skb */
+		if (unlikely(!ring->skb || err == -ENXIO)) {
 			goto out;
-
-		recv_bds += bnum;
-		clean_count += bnum;
-		if (unlikely(err)) {  /* Do jump the err */
+		} else if (likely(!err)) {
+			rx_fn(ring, ring->skb);
 			recv_pkts++;
-			continue;
 		}
 
-		/* Do update ip stack process */
-		skb->protocol = eth_type_trans(skb, netdev);
-		rx_fn(ring, skb);
-
-		recv_pkts++;
+		unused_count += ring->pending_buf;
+		ring->skb = NULL;
+		ring->pending_buf = 0;
 	}
 
 out:
-	/* Make all data has been write before submit */
-	if (clean_count + unused_count > 0)
-		hns3_nic_alloc_rx_buffers(ring,
-					  clean_count + unused_count);
-
-	return recv_pkts;
+	return failure ? budget : recv_pkts;
 }
 
-static bool hns3_get_new_int_gl(struct hns3_enet_ring_group *ring_group)
+static bool hns3_get_new_flow_lvl(struct hns3_enet_ring_group *ring_group)
 {
-	struct hns3_enet_tqp_vector *tqp_vector =
-					ring_group->ring->tqp_vector;
+#define HNS3_RX_LOW_BYTE_RATE 10000
+#define HNS3_RX_MID_BYTE_RATE 20000
+#define HNS3_RX_ULTRA_PACKET_RATE 40
+
 	enum hns3_flow_level_range new_flow_level;
-	int packets_per_msecs;
-	int bytes_per_msecs;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int packets_per_msecs, bytes_per_msecs;
 	u32 time_passed_ms;
-	u16 new_int_gl;
 
-	if (!tqp_vector->last_jiffies)
-		return false;
-
-	if (ring_group->total_packets == 0) {
-		ring_group->coal.int_gl = HNS3_INT_GL_50K;
-		ring_group->coal.flow_level = HNS3_FLOW_LOW;
-		return true;
-	}
-
-	/* Simple throttlerate management
-	 * 0-10MB/s   lower     (50000 ints/s)
-	 * 10-20MB/s   middle    (20000 ints/s)
-	 * 20-1249MB/s high      (18000 ints/s)
-	 * > 40000pps  ultra     (8000 ints/s)
-	 */
-	new_flow_level = ring_group->coal.flow_level;
-	new_int_gl = ring_group->coal.int_gl;
+	tqp_vector = ring_group->ring->tqp_vector;
 	time_passed_ms =
 		jiffies_to_msecs(jiffies - tqp_vector->last_jiffies);
-
 	if (!time_passed_ms)
 		return false;
 
@@ -2404,9 +3284,14 @@
 	do_div(ring_group->total_bytes, time_passed_ms);
 	bytes_per_msecs = ring_group->total_bytes;
 
-#define HNS3_RX_LOW_BYTE_RATE 10000
-#define HNS3_RX_MID_BYTE_RATE 20000
+	new_flow_level = ring_group->coal.flow_level;
 
+	/* Simple throttlerate management
+	 * 0-10MB/s   lower     (50000 ints/s)
+	 * 10-20MB/s   middle    (20000 ints/s)
+	 * 20-1249MB/s high      (18000 ints/s)
+	 * > 40000pps  ultra     (8000 ints/s)
+	 */
 	switch (new_flow_level) {
 	case HNS3_FLOW_LOW:
 		if (bytes_per_msecs > HNS3_RX_LOW_BYTE_RATE)
@@ -2426,13 +3311,40 @@
 		break;
 	}
 
-#define HNS3_RX_ULTRA_PACKET_RATE 40
-
 	if (packets_per_msecs > HNS3_RX_ULTRA_PACKET_RATE &&
 	    &tqp_vector->rx_group == ring_group)
 		new_flow_level = HNS3_FLOW_ULTRA;
 
-	switch (new_flow_level) {
+	ring_group->total_bytes = 0;
+	ring_group->total_packets = 0;
+	ring_group->coal.flow_level = new_flow_level;
+
+	return true;
+}
+
+static bool hns3_get_new_int_gl(struct hns3_enet_ring_group *ring_group)
+{
+	struct hns3_enet_tqp_vector *tqp_vector;
+	u16 new_int_gl;
+
+	if (!ring_group->ring)
+		return false;
+
+	tqp_vector = ring_group->ring->tqp_vector;
+	if (!tqp_vector->last_jiffies)
+		return false;
+
+	if (ring_group->total_packets == 0) {
+		ring_group->coal.int_gl = HNS3_INT_GL_50K;
+		ring_group->coal.flow_level = HNS3_FLOW_LOW;
+		return true;
+	}
+
+	if (!hns3_get_new_flow_lvl(ring_group))
+		return false;
+
+	new_int_gl = ring_group->coal.int_gl;
+	switch (ring_group->coal.flow_level) {
 	case HNS3_FLOW_LOW:
 		new_int_gl = HNS3_INT_GL_50K;
 		break;
@@ -2449,9 +3361,6 @@
 		break;
 	}
 
-	ring_group->total_bytes = 0;
-	ring_group->total_packets = 0;
-	ring_group->coal.flow_level = new_flow_level;
 	if (new_int_gl != ring_group->coal.int_gl) {
 		ring_group->coal.int_gl = new_int_gl;
 		return true;
@@ -2465,10 +3374,10 @@
 	struct hns3_enet_ring_group *tx_group = &tqp_vector->tx_group;
 	bool rx_update, tx_update;
 
-	if (tqp_vector->int_adapt_down > 0) {
-		tqp_vector->int_adapt_down--;
+	/* update param every 1000ms */
+	if (time_before(jiffies,
+			tqp_vector->last_jiffies + msecs_to_jiffies(1000)))
 		return;
-	}
 
 	if (rx_group->coal.gl_adapt_enable) {
 		rx_update = hns3_get_new_int_gl(rx_group);
@@ -2478,36 +3387,40 @@
 	}
 
 	if (tx_group->coal.gl_adapt_enable) {
-		tx_update = hns3_get_new_int_gl(&tqp_vector->tx_group);
+		tx_update = hns3_get_new_int_gl(tx_group);
 		if (tx_update)
 			hns3_set_vector_coalesce_tx_gl(tqp_vector,
 						       tx_group->coal.int_gl);
 	}
 
 	tqp_vector->last_jiffies = jiffies;
-	tqp_vector->int_adapt_down = HNS3_INT_ADAPT_DOWN_START;
 }
 
 static int hns3_nic_common_poll(struct napi_struct *napi, int budget)
 {
+	struct hns3_nic_priv *priv = netdev_priv(napi->dev);
 	struct hns3_enet_ring *ring;
 	int rx_pkt_total = 0;
 
 	struct hns3_enet_tqp_vector *tqp_vector =
 		container_of(napi, struct hns3_enet_tqp_vector, napi);
 	bool clean_complete = true;
-	int rx_budget;
+	int rx_budget = budget;
+
+	if (unlikely(test_bit(HNS3_NIC_STATE_DOWN, &priv->state))) {
+		napi_complete(napi);
+		return 0;
+	}
 
 	/* Since the actual Tx work is minimal, we can give the Tx a larger
 	 * budget and be more aggressive about cleaning up the Tx descriptors.
 	 */
-	hns3_for_each_ring(ring, tqp_vector->tx_group) {
-		if (!hns3_clean_tx_ring(ring, budget))
-			clean_complete = false;
-	}
+	hns3_for_each_ring(ring, tqp_vector->tx_group)
+		hns3_clean_tx_ring(ring, budget);
 
 	/* make sure rx ring budget not smaller than 1 */
-	rx_budget = max(budget / tqp_vector->num_tqps, 1);
+	if (tqp_vector->num_tqps > 1)
+		rx_budget = max(budget / tqp_vector->num_tqps, 1);
 
 	hns3_for_each_ring(ring, tqp_vector->rx_group) {
 		int rx_cleaned = hns3_clean_rx_ring(ring, rx_budget,
@@ -2524,9 +3437,11 @@
 	if (!clean_complete)
 		return budget;
 
-	napi_complete(napi);
-	hns3_update_new_int_gl(tqp_vector);
-	hns3_mask_vector_irq(tqp_vector, 1);
+	if (napi_complete(napi) &&
+	    likely(!test_bit(HNS3_NIC_STATE_DOWN, &priv->state))) {
+		hns3_update_new_int_gl(tqp_vector);
+		hns3_mask_vector_irq(tqp_vector, 1);
+	}
 
 	return rx_pkt_total;
 }
@@ -2638,12 +3553,31 @@
 	group->count++;
 }
 
+static void hns3_nic_set_cpumask(struct hns3_nic_priv *priv)
+{
+	struct pci_dev *pdev = priv->ae_handle->pdev;
+	struct hns3_enet_tqp_vector *tqp_vector;
+	int num_vectors = priv->vector_num;
+	int numa_node;
+	int vector_i;
+
+	numa_node = dev_to_node(&pdev->dev);
+
+	for (vector_i = 0; vector_i < num_vectors; vector_i++) {
+		tqp_vector = &priv->tqp_vector[vector_i];
+		cpumask_set_cpu(cpumask_local_spread(vector_i, numa_node),
+				&tqp_vector->affinity_mask);
+	}
+}
+
 static int hns3_nic_init_vector_data(struct hns3_nic_priv *priv)
 {
 	struct hnae3_handle *h = priv->ae_handle;
 	struct hns3_enet_tqp_vector *tqp_vector;
-	int ret = 0;
+	int ret;
 	int i;
+
+	hns3_nic_set_cpumask(priv);
 
 	for (i = 0; i < priv->vector_num; i++) {
 		tqp_vector = &priv->tqp_vector[i];
@@ -2658,13 +3592,13 @@
 		tqp_vector = &priv->tqp_vector[vector_i];
 
 		hns3_add_ring_to_group(&tqp_vector->tx_group,
-				       priv->ring_data[i].ring);
+				       &priv->ring[i]);
 
 		hns3_add_ring_to_group(&tqp_vector->rx_group,
-				       priv->ring_data[i + tqp_num].ring);
+				       &priv->ring[i + tqp_num]);
 
-		priv->ring_data[i].ring->tqp_vector = tqp_vector;
-		priv->ring_data[i + tqp_num].ring->tqp_vector = tqp_vector;
+		priv->ring[i].tqp_vector = tqp_vector;
+		priv->ring[i + tqp_num].tqp_vector = tqp_vector;
 		tqp_vector->num_tqps++;
 	}
 
@@ -2728,6 +3662,7 @@
 	if (!vector)
 		return -ENOMEM;
 
+	/* save the actual available vector number */
 	vector_num = h->ae_algo->ops->get_vector(h, vector_num, vector);
 
 	priv->vector_num = vector_num;
@@ -2758,46 +3693,38 @@
 	group->count = 0;
 }
 
-static int hns3_nic_uninit_vector_data(struct hns3_nic_priv *priv)
+static void hns3_nic_uninit_vector_data(struct hns3_nic_priv *priv)
 {
 	struct hnae3_ring_chain_node vector_ring_chain;
 	struct hnae3_handle *h = priv->ae_handle;
 	struct hns3_enet_tqp_vector *tqp_vector;
-	int i, ret;
+	int i;
 
 	for (i = 0; i < priv->vector_num; i++) {
 		tqp_vector = &priv->tqp_vector[i];
 
-		ret = hns3_get_vector_ring_chain(tqp_vector,
-						 &vector_ring_chain);
-		if (ret)
-			return ret;
+		if (!tqp_vector->rx_group.ring && !tqp_vector->tx_group.ring)
+			continue;
 
-		ret = h->ae_algo->ops->unmap_ring_from_vector(h,
+		/* Since the mapping can be overwritten, when fail to get the
+		 * chain between vector and ring, we should go on to deal with
+		 * the remaining options.
+		 */
+		if (hns3_get_vector_ring_chain(tqp_vector, &vector_ring_chain))
+			dev_warn(priv->dev, "failed to get ring chain\n");
+
+		h->ae_algo->ops->unmap_ring_from_vector(h,
 			tqp_vector->vector_irq, &vector_ring_chain);
-		if (ret)
-			return ret;
 
 		hns3_free_vector_ring_chain(tqp_vector, &vector_ring_chain);
 
-		if (tqp_vector->irq_init_flag == HNS3_VECTOR_INITED) {
-			irq_set_affinity_notifier(tqp_vector->vector_irq,
-						  NULL);
-			irq_set_affinity_hint(tqp_vector->vector_irq, NULL);
-			free_irq(tqp_vector->vector_irq, tqp_vector);
-			tqp_vector->irq_init_flag = HNS3_VECTOR_NOT_INITED;
-		}
-
-		priv->ring_data[i].ring->irq_init_flag = HNS3_VECTOR_NOT_INITED;
 		hns3_clear_ring_group(&tqp_vector->rx_group);
 		hns3_clear_ring_group(&tqp_vector->tx_group);
 		netif_napi_del(&priv->tqp_vector[i].napi);
 	}
-
-	return 0;
 }
 
-static int hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
+static void hns3_nic_dealloc_vector_data(struct hns3_nic_priv *priv)
 {
 	struct hnae3_handle *h = priv->ae_handle;
 	struct pci_dev *pdev = h->pdev;
@@ -2809,33 +3736,27 @@
 		tqp_vector = &priv->tqp_vector[i];
 		ret = h->ae_algo->ops->put_vector(h, tqp_vector->vector_irq);
 		if (ret)
-			return ret;
+			return;
 	}
 
 	devm_kfree(&pdev->dev, priv->tqp_vector);
-	return 0;
 }
 
-static int hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv,
-			     int ring_type)
+static void hns3_ring_get_cfg(struct hnae3_queue *q, struct hns3_nic_priv *priv,
+			      unsigned int ring_type)
 {
-	struct hns3_nic_ring_data *ring_data = priv->ring_data;
 	int queue_num = priv->ae_handle->kinfo.num_tqps;
-	struct pci_dev *pdev = priv->ae_handle->pdev;
 	struct hns3_enet_ring *ring;
-
-	ring = devm_kzalloc(&pdev->dev, sizeof(*ring), GFP_KERNEL);
-	if (!ring)
-		return -ENOMEM;
+	int desc_num;
 
 	if (ring_type == HNAE3_RING_TYPE_TX) {
-		ring_data[q->tqp_index].ring = ring;
-		ring_data[q->tqp_index].queue_index = q->tqp_index;
-		ring->io_base = (u8 __iomem *)q->io_base + HNS3_TX_REG_OFFSET;
+		ring = &priv->ring[q->tqp_index];
+		desc_num = priv->ae_handle->kinfo.num_tx_desc;
+		ring->queue_index = q->tqp_index;
 	} else {
-		ring_data[q->tqp_index + queue_num].ring = ring;
-		ring_data[q->tqp_index + queue_num].queue_index = q->tqp_index;
-		ring->io_base = q->io_base;
+		ring = &priv->ring[q->tqp_index + queue_num];
+		desc_num = priv->ae_handle->kinfo.num_rx_desc;
+		ring->queue_index = q->tqp_index;
 	}
 
 	hnae3_set_bit(ring->flag, HNAE3_RING_TYPE_B, ring_type);
@@ -2846,74 +3767,45 @@
 	ring->dev = priv->dev;
 	ring->desc_dma_addr = 0;
 	ring->buf_size = q->buf_size;
-	ring->desc_num = q->desc_num;
+	ring->desc_num = desc_num;
 	ring->next_to_use = 0;
 	ring->next_to_clean = 0;
-
-	return 0;
+	ring->last_to_use = 0;
 }
 
-static int hns3_queue_to_ring(struct hnae3_queue *tqp,
-			      struct hns3_nic_priv *priv)
+static void hns3_queue_to_ring(struct hnae3_queue *tqp,
+			       struct hns3_nic_priv *priv)
 {
-	int ret;
-
-	ret = hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_TX);
-	if (ret)
-		return ret;
-
-	ret = hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_RX);
-	if (ret) {
-		devm_kfree(priv->dev, priv->ring_data[tqp->tqp_index].ring);
-		return ret;
-	}
-
-	return 0;
+	hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_TX);
+	hns3_ring_get_cfg(tqp, priv, HNAE3_RING_TYPE_RX);
 }
 
 static int hns3_get_ring_config(struct hns3_nic_priv *priv)
 {
 	struct hnae3_handle *h = priv->ae_handle;
 	struct pci_dev *pdev = h->pdev;
-	int i, ret;
+	int i;
 
-	priv->ring_data =  devm_kzalloc(&pdev->dev,
-					array3_size(h->kinfo.num_tqps,
-						    sizeof(*priv->ring_data),
-						    2),
-					GFP_KERNEL);
-	if (!priv->ring_data)
+	priv->ring = devm_kzalloc(&pdev->dev,
+				  array3_size(h->kinfo.num_tqps,
+					      sizeof(*priv->ring), 2),
+				  GFP_KERNEL);
+	if (!priv->ring)
 		return -ENOMEM;
 
-	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		ret = hns3_queue_to_ring(h->kinfo.tqp[i], priv);
-		if (ret)
-			goto err;
-	}
+	for (i = 0; i < h->kinfo.num_tqps; i++)
+		hns3_queue_to_ring(h->kinfo.tqp[i], priv);
 
 	return 0;
-err:
-	while (i--) {
-		devm_kfree(priv->dev, priv->ring_data[i].ring);
-		devm_kfree(priv->dev,
-			   priv->ring_data[i + h->kinfo.num_tqps].ring);
-	}
-
-	devm_kfree(&pdev->dev, priv->ring_data);
-	return ret;
 }
 
 static void hns3_put_ring_config(struct hns3_nic_priv *priv)
 {
-	struct hnae3_handle *h = priv->ae_handle;
-	int i;
+	if (!priv->ring)
+		return;
 
-	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		devm_kfree(priv->dev, priv->ring_data[i].ring);
-		devm_kfree(priv->dev,
-			   priv->ring_data[i + h->kinfo.num_tqps].ring);
-	}
-	devm_kfree(priv->dev, priv->ring_data);
+	devm_kfree(priv->dev, priv->ring);
+	priv->ring = NULL;
 }
 
 static int hns3_alloc_ring_memory(struct hns3_enet_ring *ring)
@@ -2923,8 +3815,8 @@
 	if (ring->desc_num <= 0 || ring->buf_size <= 0)
 		return -EINVAL;
 
-	ring->desc_cb = kcalloc(ring->desc_num, sizeof(ring->desc_cb[0]),
-				GFP_KERNEL);
+	ring->desc_cb = devm_kcalloc(ring_to_dev(ring), ring->desc_num,
+				     sizeof(ring->desc_cb[0]), GFP_KERNEL);
 	if (!ring->desc_cb) {
 		ret = -ENOMEM;
 		goto out;
@@ -2945,19 +3837,25 @@
 out_with_desc:
 	hns3_free_desc(ring);
 out_with_desc_cb:
-	kfree(ring->desc_cb);
+	devm_kfree(ring_to_dev(ring), ring->desc_cb);
 	ring->desc_cb = NULL;
 out:
 	return ret;
 }
 
-static void hns3_fini_ring(struct hns3_enet_ring *ring)
+void hns3_fini_ring(struct hns3_enet_ring *ring)
 {
 	hns3_free_desc(ring);
-	kfree(ring->desc_cb);
+	devm_kfree(ring_to_dev(ring), ring->desc_cb);
 	ring->desc_cb = NULL;
 	ring->next_to_clean = 0;
 	ring->next_to_use = 0;
+	ring->last_to_use = 0;
+	ring->pending_buf = 0;
+	if (ring->skb) {
+		dev_kfree_skb_any(ring->skb);
+		ring->skb = NULL;
+	}
 }
 
 static int hns3_buf_size2type(u32 buf_size)
@@ -2990,8 +3888,7 @@
 	struct hnae3_queue *q = ring->tqp;
 
 	if (!HNAE3_IS_TX_RING(ring)) {
-		hns3_write_dev(q, HNS3_RING_RX_RING_BASEADDR_L_REG,
-			       (u32)dma);
+		hns3_write_dev(q, HNS3_RING_RX_RING_BASEADDR_L_REG, (u32)dma);
 		hns3_write_dev(q, HNS3_RING_RX_RING_BASEADDR_H_REG,
 			       (u32)((dma >> 31) >> 1));
 
@@ -3026,7 +3923,7 @@
 		for (j = 0; j < tc_info->tqp_count; j++) {
 			struct hnae3_queue *q;
 
-			q = priv->ring_data[tc_info->tqp_offset + j].ring->tqp;
+			q = priv->ring[tc_info->tqp_offset + j].tqp;
 			hns3_write_dev(q, HNS3_RING_TX_RING_TC_REG,
 				       tc_info->tc);
 		}
@@ -3041,21 +3938,21 @@
 	int ret;
 
 	for (i = 0; i < ring_num; i++) {
-		ret = hns3_alloc_ring_memory(priv->ring_data[i].ring);
+		ret = hns3_alloc_ring_memory(&priv->ring[i]);
 		if (ret) {
 			dev_err(priv->dev,
 				"Alloc ring memory fail! ret=%d\n", ret);
 			goto out_when_alloc_ring_memory;
 		}
 
-		u64_stats_init(&priv->ring_data[i].ring->syncp);
+		u64_stats_init(&priv->ring[i].syncp);
 	}
 
 	return 0;
 
 out_when_alloc_ring_memory:
 	for (j = i - 1; j >= 0; j--)
-		hns3_fini_ring(priv->ring_data[j].ring);
+		hns3_fini_ring(&priv->ring[j]);
 
 	return -ENOMEM;
 }
@@ -3066,71 +3963,111 @@
 	int i;
 
 	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		if (h->ae_algo->ops->reset_queue)
-			h->ae_algo->ops->reset_queue(h, i);
-
-		hns3_fini_ring(priv->ring_data[i].ring);
-		hns3_fini_ring(priv->ring_data[i + h->kinfo.num_tqps].ring);
+		hns3_fini_ring(&priv->ring[i]);
+		hns3_fini_ring(&priv->ring[i + h->kinfo.num_tqps]);
 	}
 	return 0;
 }
 
 /* Set mac addr if it is configured. or leave it to the AE driver */
-static void hns3_init_mac_addr(struct net_device *netdev, bool init)
+static int hns3_init_mac_addr(struct net_device *netdev)
 {
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	struct hnae3_handle *h = priv->ae_handle;
 	u8 mac_addr_temp[ETH_ALEN];
+	int ret = 0;
 
-	if (h->ae_algo->ops->get_mac_addr && init) {
+	if (h->ae_algo->ops->get_mac_addr)
 		h->ae_algo->ops->get_mac_addr(h, mac_addr_temp);
-		ether_addr_copy(netdev->dev_addr, mac_addr_temp);
-	}
 
 	/* Check if the MAC address is valid, if not get a random one */
-	if (!is_valid_ether_addr(netdev->dev_addr)) {
+	if (!is_valid_ether_addr(mac_addr_temp)) {
 		eth_hw_addr_random(netdev);
 		dev_warn(priv->dev, "using random MAC address %pM\n",
 			 netdev->dev_addr);
+	} else if (!ether_addr_equal(netdev->dev_addr, mac_addr_temp)) {
+		ether_addr_copy(netdev->dev_addr, mac_addr_temp);
+		ether_addr_copy(netdev->perm_addr, mac_addr_temp);
+	} else {
+		return 0;
 	}
 
 	if (h->ae_algo->ops->set_mac_addr)
-		h->ae_algo->ops->set_mac_addr(h, netdev->dev_addr, true);
+		ret = h->ae_algo->ops->set_mac_addr(h, netdev->dev_addr, true);
 
+	return ret;
 }
 
-static void hns3_uninit_mac_addr(struct net_device *netdev)
+static int hns3_init_phy(struct net_device *netdev)
 {
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
-	struct hnae3_handle *h = priv->ae_handle;
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+	int ret = 0;
 
-	if (h->ae_algo->ops->rm_uc_addr)
-		h->ae_algo->ops->rm_uc_addr(h, netdev->dev_addr);
+	if (h->ae_algo->ops->mac_connect_phy)
+		ret = h->ae_algo->ops->mac_connect_phy(h);
+
+	return ret;
 }
 
-static void hns3_nic_set_priv_ops(struct net_device *netdev)
+static void hns3_uninit_phy(struct net_device *netdev)
 {
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
+	struct hnae3_handle *h = hns3_get_handle(netdev);
 
-	if ((netdev->features & NETIF_F_TSO) ||
-	    (netdev->features & NETIF_F_TSO6)) {
-		priv->ops.fill_desc = hns3_fill_desc_tso;
-		priv->ops.maybe_stop_tx = hns3_nic_maybe_stop_tso;
-	} else {
-		priv->ops.fill_desc = hns3_fill_desc;
-		priv->ops.maybe_stop_tx = hns3_nic_maybe_stop_tx;
-	}
+	if (h->ae_algo->ops->mac_disconnect_phy)
+		h->ae_algo->ops->mac_disconnect_phy(h);
+}
+
+static void hns3_del_all_fd_rules(struct net_device *netdev, bool clear_list)
+{
+	struct hnae3_handle *h = hns3_get_handle(netdev);
+
+	if (h->ae_algo->ops->del_all_fd_entries)
+		h->ae_algo->ops->del_all_fd_entries(h, clear_list);
+}
+
+static int hns3_client_start(struct hnae3_handle *handle)
+{
+	if (!handle->ae_algo->ops->client_start)
+		return 0;
+
+	return handle->ae_algo->ops->client_start(handle);
+}
+
+static void hns3_client_stop(struct hnae3_handle *handle)
+{
+	if (!handle->ae_algo->ops->client_stop)
+		return;
+
+	handle->ae_algo->ops->client_stop(handle);
+}
+
+static void hns3_info_show(struct hns3_nic_priv *priv)
+{
+	struct hnae3_knic_private_info *kinfo = &priv->ae_handle->kinfo;
+
+	dev_info(priv->dev, "MAC address: %pM\n", priv->netdev->dev_addr);
+	dev_info(priv->dev, "Task queue pairs numbers: %u\n", kinfo->num_tqps);
+	dev_info(priv->dev, "RSS size: %u\n", kinfo->rss_size);
+	dev_info(priv->dev, "Allocated RSS size: %u\n", kinfo->req_rss_size);
+	dev_info(priv->dev, "RX buffer length: %u\n", kinfo->rx_buf_len);
+	dev_info(priv->dev, "Desc num per TX queue: %u\n", kinfo->num_tx_desc);
+	dev_info(priv->dev, "Desc num per RX queue: %u\n", kinfo->num_rx_desc);
+	dev_info(priv->dev, "Total number of enabled TCs: %u\n", kinfo->num_tc);
+	dev_info(priv->dev, "Max mtu size: %u\n", priv->netdev->max_mtu);
 }
 
 static int hns3_client_init(struct hnae3_handle *handle)
 {
 	struct pci_dev *pdev = handle->pdev;
+	struct hnae3_ae_dev *ae_dev = pci_get_drvdata(pdev);
+	u16 alloc_tqps, max_rss_size;
 	struct hns3_nic_priv *priv;
 	struct net_device *netdev;
 	int ret;
 
-	netdev = alloc_etherdev_mq(sizeof(struct hns3_nic_priv),
-				   hns3_get_max_available_channels(handle));
+	handle->ae_algo->ops->get_tqps_and_rss_info(handle, &alloc_tqps,
+						    &max_rss_size);
+	netdev = alloc_etherdev_mq(sizeof(struct hns3_nic_priv), alloc_tqps);
 	if (!netdev)
 		return -ENOMEM;
 
@@ -3138,13 +4075,16 @@
 	priv->dev = &pdev->dev;
 	priv->netdev = netdev;
 	priv->ae_handle = handle;
-	priv->ae_handle->last_reset_time = jiffies;
 	priv->tx_timeout_count = 0;
+	priv->max_non_tso_bd_num = ae_dev->dev_specs.max_non_tso_bd_num;
+	set_bit(HNS3_NIC_STATE_DOWN, &priv->state);
+
+	handle->msg_enable = netif_msg_init(debug, DEFAULT_MSG_LEVEL);
 
 	handle->kinfo.netdev = netdev;
 	handle->priv = (void *)priv;
 
-	hns3_init_mac_addr(netdev, true);
+	hns3_init_mac_addr(netdev);
 
 	hns3_set_default_feature(netdev);
 
@@ -3153,15 +4093,9 @@
 	netdev->netdev_ops = &hns3_nic_netdev_ops;
 	SET_NETDEV_DEV(netdev, &pdev->dev);
 	hns3_ethtool_set_ops(netdev);
-	hns3_nic_set_priv_ops(netdev);
 
 	/* Carrier off reporting is important to ethtool even BEFORE open */
 	netif_carrier_off(netdev);
-
-	if (handle->flags & HNAE3_SUPPORT_VF)
-		handle->reset_level = HNAE3_VF_RESET;
-	else
-		handle->reset_level = HNAE3_FUNC_RESET;
 
 	ret = hns3_get_ring_config(priv);
 	if (ret) {
@@ -3184,8 +4118,39 @@
 	ret = hns3_init_all_ring(priv);
 	if (ret) {
 		ret = -ENOMEM;
-		goto out_init_ring_data;
+		goto out_init_ring;
 	}
+
+	ret = hns3_init_phy(netdev);
+	if (ret)
+		goto out_init_phy;
+
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
+	if (ret) {
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto out_init_irq_fail;
+	}
+
+	ret = hns3_client_start(handle);
+	if (ret) {
+		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
+		goto out_client_start;
+	}
+
+	hns3_dcbnl_setup(handle);
+
+	hns3_dbg_init(handle);
+
+	/* MTU range: (ETH_MIN_MTU(kernel default) - 9702) */
+	netdev->max_mtu = HNS3_MAX_MTU;
+
+	set_bit(HNS3_NIC_STATE_INITED, &priv->state);
 
 	ret = register_netdev(netdev);
 	if (ret) {
@@ -3193,20 +4158,26 @@
 		goto out_reg_netdev_fail;
 	}
 
-	hns3_dcbnl_setup(handle);
-
-	/* MTU range: (ETH_MIN_MTU(kernel default) - 9706) */
-	netdev->max_mtu = HNS3_MAX_MTU - (ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN);
+	if (netif_msg_drv(handle))
+		hns3_info_show(priv);
 
 	return ret;
 
 out_reg_netdev_fail:
-out_init_ring_data:
-	(void)hns3_nic_uninit_vector_data(priv);
+	hns3_dbg_uninit(handle);
+out_client_start:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+out_init_irq_fail:
+	hns3_uninit_phy(netdev);
+out_init_phy:
+	hns3_uninit_all_ring(priv);
+out_init_ring:
+	hns3_nic_uninit_vector_data(priv);
 out_init_vector_data:
 	hns3_nic_dealloc_vector_data(priv);
 out_alloc_vector_data:
-	priv->ring_data = NULL;
+	priv->ring = NULL;
 out_get_ring_cfg:
 	priv->ae_handle = NULL;
 	free_netdev(netdev);
@@ -3222,15 +4193,26 @@
 	if (netdev->reg_state != NETREG_UNINITIALIZED)
 		unregister_netdev(netdev);
 
-	hns3_force_clear_all_rx_ring(handle);
+	hns3_client_stop(handle);
 
-	ret = hns3_nic_uninit_vector_data(priv);
-	if (ret)
-		netdev_err(netdev, "uninit vector error\n");
+	hns3_uninit_phy(netdev);
 
-	ret = hns3_nic_dealloc_vector_data(priv);
-	if (ret)
-		netdev_err(netdev, "dealloc vector error\n");
+	if (!test_and_clear_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_warn(netdev, "already uninitialized\n");
+		goto out_netdev_free;
+	}
+
+	hns3_free_rx_cpu_rmap(netdev);
+
+	hns3_nic_uninit_irq(priv);
+
+	hns3_del_all_fd_rules(netdev, true);
+
+	hns3_clear_all_ring(handle, true);
+
+	hns3_nic_uninit_vector_data(priv);
+
+	hns3_nic_dealloc_vector_data(priv);
 
 	ret = hns3_uninit_all_ring(priv);
 	if (ret)
@@ -3238,10 +4220,8 @@
 
 	hns3_put_ring_config(priv);
 
-	priv->ring_data = NULL;
-
-	hns3_uninit_mac_addr(netdev);
-
+out_netdev_free:
+	hns3_dbg_uninit(handle);
 	free_netdev(netdev);
 }
 
@@ -3253,13 +4233,15 @@
 		return;
 
 	if (linkup) {
-		netif_carrier_on(netdev);
 		netif_tx_wake_all_queues(netdev);
-		netdev_info(netdev, "link up\n");
+		netif_carrier_on(netdev);
+		if (netif_msg_link(handle))
+			netdev_info(netdev, "link up\n");
 	} else {
 		netif_carrier_off(netdev);
 		netif_tx_stop_all_queues(netdev);
-		netdev_info(netdev, "link down\n");
+		if (netif_msg_link(handle))
+			netdev_info(netdev, "link down\n");
 	}
 }
 
@@ -3267,8 +4249,6 @@
 {
 	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
 	struct net_device *ndev = kinfo->netdev;
-	bool if_running;
-	int ret;
 
 	if (tc > HNAE3_MAX_TC)
 		return -EINVAL;
@@ -3276,50 +4256,18 @@
 	if (!ndev)
 		return -ENODEV;
 
-	if_running = netif_running(ndev);
-
-	if (if_running) {
-		(void)hns3_nic_net_stop(ndev);
-		msleep(100);
-	}
-
-	ret = (kinfo->dcb_ops && kinfo->dcb_ops->map_update) ?
-		kinfo->dcb_ops->map_update(handle) : -EOPNOTSUPP;
-	if (ret)
-		goto err_out;
-
-	ret = hns3_nic_set_real_num_queue(ndev);
-
-err_out:
-	if (if_running)
-		(void)hns3_nic_net_open(ndev);
-
-	return ret;
-}
-
-static void hns3_recover_hw_addr(struct net_device *ndev)
-{
-	struct netdev_hw_addr_list *list;
-	struct netdev_hw_addr *ha, *tmp;
-
-	/* go through and sync uc_addr entries to the device */
-	list = &ndev->uc;
-	list_for_each_entry_safe(ha, tmp, &list->list, list)
-		hns3_nic_uc_sync(ndev, ha->addr);
-
-	/* go through and sync mc_addr entries to the device */
-	list = &ndev->mc;
-	list_for_each_entry_safe(ha, tmp, &list->list, list)
-		hns3_nic_mc_sync(ndev, ha->addr);
+	return hns3_nic_set_real_num_queue(ndev);
 }
 
 static void hns3_clear_tx_ring(struct hns3_enet_ring *ring)
 {
 	while (ring->next_to_clean != ring->next_to_use) {
 		ring->desc[ring->next_to_clean].tx.bdtp_fe_sc_vld_ra_ri = 0;
-		hns3_free_buffer_detach(ring, ring->next_to_clean);
+		hns3_free_buffer_detach(ring, ring->next_to_clean, 0);
 		ring_ptr_move_fw(ring, next_to_clean);
 	}
+
+	ring->pending_buf = 0;
 }
 
 static int hns3_clear_rx_ring(struct hns3_enet_ring *ring)
@@ -3333,7 +4281,7 @@
 		 * stack, so we need to replace the buffer here.
 		 */
 		if (!ring->desc_cb[ring->next_to_use].reuse_flag) {
-			ret = hns3_reserve_buffer_map(ring, &res_cbs);
+			ret = hns3_alloc_and_map_buffer(ring, &res_cbs);
 			if (ret) {
 				u64_stats_update_begin(&ring->syncp);
 				ring->stats.sw_err_cnt++;
@@ -3341,15 +4289,21 @@
 				/* if alloc new buffer fail, exit directly
 				 * and reclear in up flow.
 				 */
-				netdev_warn(ring->tqp->handle->kinfo.netdev,
+				netdev_warn(ring_to_netdev(ring),
 					    "reserve buffer map failed, ret = %d\n",
 					    ret);
 				return ret;
 			}
-			hns3_replace_buffer(ring, ring->next_to_use,
-					    &res_cbs);
+			hns3_replace_buffer(ring, ring->next_to_use, &res_cbs);
 		}
 		ring_ptr_move_fw(ring, next_to_use);
+	}
+
+	/* Free the pending skb in rx ring */
+	if (ring->skb) {
+		dev_kfree_skb_any(ring->skb);
+		ring->skb = NULL;
+		ring->pending_buf = 0;
 	}
 
 	return 0;
@@ -3372,40 +4326,26 @@
 	}
 }
 
-static void hns3_force_clear_all_rx_ring(struct hnae3_handle *h)
-{
-	struct net_device *ndev = h->kinfo.netdev;
-	struct hns3_nic_priv *priv = netdev_priv(ndev);
-	struct hns3_enet_ring *ring;
-	u32 i;
-
-	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		ring = priv->ring_data[i + h->kinfo.num_tqps].ring;
-		hns3_force_clear_rx_ring(ring);
-	}
-}
-
-static void hns3_clear_all_ring(struct hnae3_handle *h)
+static void hns3_clear_all_ring(struct hnae3_handle *h, bool force)
 {
 	struct net_device *ndev = h->kinfo.netdev;
 	struct hns3_nic_priv *priv = netdev_priv(ndev);
 	u32 i;
 
 	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		struct netdev_queue *dev_queue;
 		struct hns3_enet_ring *ring;
 
-		ring = priv->ring_data[i].ring;
+		ring = &priv->ring[i];
 		hns3_clear_tx_ring(ring);
-		dev_queue = netdev_get_tx_queue(ndev,
-						priv->ring_data[i].queue_index);
-		netdev_tx_reset_queue(dev_queue);
 
-		ring = priv->ring_data[i + h->kinfo.num_tqps].ring;
+		ring = &priv->ring[i + h->kinfo.num_tqps];
 		/* Continue to clear other rings even if clearing some
 		 * rings failed.
 		 */
-		hns3_clear_rx_ring(ring);
+		if (force)
+			hns3_force_clear_rx_ring(ring);
+		else
+			hns3_clear_rx_ring(ring);
 	}
 }
 
@@ -3418,17 +4358,21 @@
 	int ret;
 
 	for (i = 0; i < h->kinfo.num_tqps; i++) {
-		h->ae_algo->ops->reset_queue(h, i);
-		hns3_init_ring_hw(priv->ring_data[i].ring);
+		ret = h->ae_algo->ops->reset_queue(h, i);
+		if (ret)
+			return ret;
+
+		hns3_init_ring_hw(&priv->ring[i]);
 
 		/* We need to clear tx ring here because self test will
 		 * use the ring and will not run down before up
 		 */
-		hns3_clear_tx_ring(priv->ring_data[i].ring);
-		priv->ring_data[i].ring->next_to_clean = 0;
-		priv->ring_data[i].ring->next_to_use = 0;
+		hns3_clear_tx_ring(&priv->ring[i]);
+		priv->ring[i].next_to_clean = 0;
+		priv->ring[i].next_to_use = 0;
+		priv->ring[i].last_to_use = 0;
 
-		rx_ring = priv->ring_data[i + h->kinfo.num_tqps].ring;
+		rx_ring = &priv->ring[i + h->kinfo.num_tqps];
 		hns3_init_ring_hw(rx_ring);
 		ret = hns3_clear_rx_ring(rx_ring);
 		if (ret)
@@ -3452,8 +4396,8 @@
 static void hns3_store_coal(struct hns3_nic_priv *priv)
 {
 	/* ethtool only support setting and querying one coal
-	 * configuation for now, so save the vector 0' coal
-	 * configuation here in order to restore it.
+	 * configuration for now, so save the vector 0' coal
+	 * configuration here in order to restore it.
 	 */
 	memcpy(&priv->tx_coal, &priv->tqp_vector[0].tx_group.coal,
 	       sizeof(struct hns3_enet_coalesce));
@@ -3478,6 +4422,10 @@
 {
 	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
 	struct net_device *ndev = kinfo->netdev;
+	struct hns3_nic_priv *priv = netdev_priv(ndev);
+
+	if (test_and_set_bit(HNS3_NIC_STATE_RESETTING, &priv->state))
+		return 0;
 
 	if (!netif_running(ndev))
 		return 0;
@@ -3488,16 +4436,24 @@
 static int hns3_reset_notify_up_enet(struct hnae3_handle *handle)
 {
 	struct hnae3_knic_private_info *kinfo = &handle->kinfo;
+	struct hns3_nic_priv *priv = netdev_priv(kinfo->netdev);
 	int ret = 0;
 
+	if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_err(kinfo->netdev, "device is not initialized yet\n");
+		return -EFAULT;
+	}
+
+	clear_bit(HNS3_NIC_STATE_RESETTING, &priv->state);
+
 	if (netif_running(kinfo->netdev)) {
-		ret = hns3_nic_net_up(kinfo->netdev);
+		ret = hns3_nic_net_open(kinfo->netdev);
 		if (ret) {
+			set_bit(HNS3_NIC_STATE_RESETTING, &priv->state);
 			netdev_err(kinfo->netdev,
-				   "hns net up fail, ret=%d!\n", ret);
+				   "net up fail, ret=%d!\n", ret);
 			return ret;
 		}
-		handle->last_reset_time = jiffies;
 	}
 
 	return ret;
@@ -3509,28 +4465,63 @@
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	int ret;
 
-	hns3_init_mac_addr(netdev, false);
-	hns3_nic_set_rx_mode(netdev);
-	hns3_recover_hw_addr(netdev);
-
-	/* Hardware table is only clear when pf resets */
-	if (!(handle->flags & HNAE3_SUPPORT_VF))
-		hns3_restore_vlan(netdev);
-
 	/* Carrier off reporting is important to ethtool even BEFORE open */
 	netif_carrier_off(netdev);
+
+	ret = hns3_get_ring_config(priv);
+	if (ret)
+		return ret;
+
+	ret = hns3_nic_alloc_vector_data(priv);
+	if (ret)
+		goto err_put_ring;
 
 	hns3_restore_coal(priv);
 
 	ret = hns3_nic_init_vector_data(priv);
 	if (ret)
-		return ret;
+		goto err_dealloc_vector;
 
 	ret = hns3_init_all_ring(priv);
+	if (ret)
+		goto err_uninit_vector;
+
+	/* the device can work without cpu rmap, only aRFS needs it */
+	ret = hns3_set_rx_cpu_rmap(netdev);
+	if (ret)
+		dev_warn(priv->dev, "set rx cpu rmap fail, ret=%d\n", ret);
+
+	ret = hns3_nic_init_irq(priv);
 	if (ret) {
-		hns3_nic_uninit_vector_data(priv);
-		priv->ring_data = NULL;
+		dev_err(priv->dev, "init irq failed! ret=%d\n", ret);
+		hns3_free_rx_cpu_rmap(netdev);
+		goto err_init_irq_fail;
 	}
+
+	if (!hns3_is_phys_func(handle->pdev))
+		hns3_init_mac_addr(netdev);
+
+	ret = hns3_client_start(handle);
+	if (ret) {
+		dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret);
+		goto err_client_start_fail;
+	}
+
+	set_bit(HNS3_NIC_STATE_INITED, &priv->state);
+
+	return ret;
+
+err_client_start_fail:
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+err_init_irq_fail:
+	hns3_uninit_all_ring(priv);
+err_uninit_vector:
+	hns3_nic_uninit_vector_data(priv);
+err_dealloc_vector:
+	hns3_nic_dealloc_vector_data(priv);
+err_put_ring:
+	hns3_put_ring_config(priv);
 
 	return ret;
 }
@@ -3541,21 +4532,27 @@
 	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	int ret;
 
-	hns3_force_clear_all_rx_ring(handle);
-
-	ret = hns3_nic_uninit_vector_data(priv);
-	if (ret) {
-		netdev_err(netdev, "uninit vector error\n");
-		return ret;
+	if (!test_and_clear_bit(HNS3_NIC_STATE_INITED, &priv->state)) {
+		netdev_warn(netdev, "already uninitialized\n");
+		return 0;
 	}
 
+	hns3_free_rx_cpu_rmap(netdev);
+	hns3_nic_uninit_irq(priv);
+	hns3_clear_all_ring(handle, true);
+	hns3_reset_tx_queue(priv->ae_handle);
+
+	hns3_nic_uninit_vector_data(priv);
+
 	hns3_store_coal(priv);
+
+	hns3_nic_dealloc_vector_data(priv);
 
 	ret = hns3_uninit_all_ring(priv);
 	if (ret)
 		netdev_err(netdev, "uninit ring error\n");
 
-	hns3_uninit_mac_addr(netdev);
+	hns3_put_ring_config(priv);
 
 	return ret;
 }
@@ -3585,113 +4582,112 @@
 	return ret;
 }
 
-static int hns3_modify_tqp_num(struct net_device *netdev, u16 new_tqp_num)
+static int hns3_change_channels(struct hnae3_handle *handle, u32 new_tqp_num,
+				bool rxfh_configured)
 {
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
-	struct hnae3_handle *h = hns3_get_handle(netdev);
 	int ret;
 
-	ret = h->ae_algo->ops->set_channels(h, new_tqp_num);
+	ret = handle->ae_algo->ops->set_channels(handle, new_tqp_num,
+						 rxfh_configured);
+	if (ret) {
+		dev_err(&handle->pdev->dev,
+			"Change tqp num(%u) fail.\n", new_tqp_num);
+		return ret;
+	}
+
+	ret = hns3_reset_notify(handle, HNAE3_INIT_CLIENT);
 	if (ret)
 		return ret;
 
-	ret = hns3_get_ring_config(priv);
+	ret =  hns3_reset_notify(handle, HNAE3_UP_CLIENT);
 	if (ret)
-		return ret;
+		hns3_reset_notify(handle, HNAE3_UNINIT_CLIENT);
 
-	ret = hns3_nic_alloc_vector_data(priv);
-	if (ret)
-		goto err_alloc_vector;
-
-	hns3_restore_coal(priv);
-
-	ret = hns3_nic_init_vector_data(priv);
-	if (ret)
-		goto err_uninit_vector;
-
-	ret = hns3_init_all_ring(priv);
-	if (ret)
-		goto err_put_ring;
-
-	return 0;
-
-err_put_ring:
-	hns3_put_ring_config(priv);
-err_uninit_vector:
-	hns3_nic_uninit_vector_data(priv);
-err_alloc_vector:
-	hns3_nic_dealloc_vector_data(priv);
 	return ret;
-}
-
-static int hns3_adjust_tqps_num(u8 num_tc, u32 new_tqp_num)
-{
-	return (new_tqp_num / num_tc) * num_tc;
 }
 
 int hns3_set_channels(struct net_device *netdev,
 		      struct ethtool_channels *ch)
 {
-	struct hns3_nic_priv *priv = netdev_priv(netdev);
 	struct hnae3_handle *h = hns3_get_handle(netdev);
 	struct hnae3_knic_private_info *kinfo = &h->kinfo;
-	bool if_running = netif_running(netdev);
+	bool rxfh_configured = netif_is_rxfh_configured(netdev);
 	u32 new_tqp_num = ch->combined_count;
 	u16 org_tqp_num;
 	int ret;
+
+	if (hns3_nic_resetting(netdev))
+		return -EBUSY;
 
 	if (ch->rx_count || ch->tx_count)
 		return -EINVAL;
 
 	if (new_tqp_num > hns3_get_max_available_channels(h) ||
-	    new_tqp_num < kinfo->num_tc) {
+	    new_tqp_num < 1) {
 		dev_err(&netdev->dev,
-			"Change tqps fail, the tqp range is from %d to %d",
-			kinfo->num_tc,
+			"Change tqps fail, the tqp range is from 1 to %u",
 			hns3_get_max_available_channels(h));
 		return -EINVAL;
 	}
 
-	new_tqp_num = hns3_adjust_tqps_num(kinfo->num_tc, new_tqp_num);
-	if (kinfo->num_tqps == new_tqp_num)
+	if (kinfo->rss_size == new_tqp_num)
 		return 0;
 
-	if (if_running)
-		hns3_nic_net_stop(netdev);
+	netif_dbg(h, drv, netdev,
+		  "set channels: tqp_num=%u, rxfh=%d\n",
+		  new_tqp_num, rxfh_configured);
 
-	ret = hns3_nic_uninit_vector_data(priv);
-	if (ret) {
-		dev_err(&netdev->dev,
-			"Unbind vector with tqp fail, nothing is changed");
-		goto open_netdev;
-	}
+	ret = hns3_reset_notify(h, HNAE3_DOWN_CLIENT);
+	if (ret)
+		return ret;
 
-	hns3_store_coal(priv);
-
-	hns3_nic_dealloc_vector_data(priv);
-
-	hns3_uninit_all_ring(priv);
-	hns3_put_ring_config(priv);
+	ret = hns3_reset_notify(h, HNAE3_UNINIT_CLIENT);
+	if (ret)
+		return ret;
 
 	org_tqp_num = h->kinfo.num_tqps;
-	ret = hns3_modify_tqp_num(netdev, new_tqp_num);
+	ret = hns3_change_channels(h, new_tqp_num, rxfh_configured);
 	if (ret) {
-		ret = hns3_modify_tqp_num(netdev, org_tqp_num);
-		if (ret) {
-			/* If revert to old tqp failed, fatal error occurred */
-			dev_err(&netdev->dev,
-				"Revert to old tqp num fail, ret=%d", ret);
-			return ret;
+		int ret1;
+
+		netdev_warn(netdev,
+			    "Change channels fail, revert to old value\n");
+		ret1 = hns3_change_channels(h, org_tqp_num, rxfh_configured);
+		if (ret1) {
+			netdev_err(netdev,
+				   "revert to old channel fail\n");
+			return ret1;
 		}
-		dev_info(&netdev->dev,
-			 "Change tqp num fail, Revert to old tqp num");
+
+		return ret;
 	}
 
-open_netdev:
-	if (if_running)
-		hns3_nic_net_open(netdev);
+	return 0;
+}
 
-	return ret;
+static const struct hns3_hw_error_info hns3_hw_err[] = {
+	{ .type = HNAE3_PPU_POISON_ERROR,
+	  .msg = "PPU poison" },
+	{ .type = HNAE3_CMDQ_ECC_ERROR,
+	  .msg = "IMP CMDQ error" },
+	{ .type = HNAE3_IMP_RD_POISON_ERROR,
+	  .msg = "IMP RD poison" },
+	{ .type = HNAE3_ROCEE_AXI_RESP_ERROR,
+	  .msg = "ROCEE AXI RESP error" },
+};
+
+static void hns3_process_hw_error(struct hnae3_handle *handle,
+				  enum hnae3_hw_error_type type)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hns3_hw_err); i++) {
+		if (hns3_hw_err[i].type == type) {
+			dev_err(&handle->pdev->dev, "Detected %s!\n",
+				hns3_hw_err[i].msg);
+			break;
+		}
+	}
 }
 
 static const struct hnae3_client_ops client_ops = {
@@ -3700,6 +4696,7 @@
 	.link_status_change = hns3_link_status_change,
 	.setup_tc = hns3_client_setup_tc,
 	.reset_notify = hns3_reset_notify,
+	.process_hw_error = hns3_process_hw_error,
 };
 
 /* hns3_init_module - Driver registration routine
@@ -3714,21 +4711,29 @@
 	pr_info("%s: %s\n", hns3_driver_name, hns3_copyright);
 
 	client.type = HNAE3_CLIENT_KNIC;
-	snprintf(client.name, HNAE3_CLIENT_NAME_LENGTH - 1, "%s",
+	snprintf(client.name, HNAE3_CLIENT_NAME_LENGTH, "%s",
 		 hns3_driver_name);
 
 	client.ops = &client_ops;
 
 	INIT_LIST_HEAD(&client.node);
 
+	hns3_dbg_register_debugfs(hns3_driver_name);
+
 	ret = hnae3_register_client(&client);
 	if (ret)
-		return ret;
+		goto err_reg_client;
 
 	ret = pci_register_driver(&hns3_driver);
 	if (ret)
-		hnae3_unregister_client(&client);
+		goto err_reg_driver;
 
+	return ret;
+
+err_reg_driver:
+	hnae3_unregister_client(&client);
+err_reg_client:
+	hns3_dbg_unregister_debugfs();
 	return ret;
 }
 module_init(hns3_init_module);
@@ -3741,6 +4746,7 @@
 {
 	pci_unregister_driver(&hns3_driver);
 	hnae3_unregister_client(&client);
+	hns3_dbg_unregister_debugfs();
 }
 module_exit(hns3_exit_module);
 
@@ -3748,4 +4754,3 @@
 MODULE_AUTHOR("Huawei Tech. Co., Ltd.");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("pci:hns-nic");
-MODULE_VERSION(HNS3_MOD_VERSION);

--
Gitblit v1.6.2