5 files changed, 436 insertions, 126 deletions

diff --git a/drivers/net/hyperv/hyperv_net.h b/drivers/net/hyperv/hyperv_net.h
index 315278a7cf88..dd5919ec408b 100644
--- a/drivers/net/hyperv/hyperv_net.h
+++ b/drivers/net/hyperv/hyperv_net.h
@@ -15,6 +15,7 @@
 #include <linux/list.h>
 #include <linux/hyperv.h>
 #include <linux/rndis.h>
+#include <linux/jhash.h>
 
 /* RSS related */
 #define OID_GEN_RECEIVE_SCALE_CAPABILITIES 0x00010203  /* query only */
@@ -164,6 +165,7 @@ struct hv_netvsc_packet {
 	u32 total_bytes;
 	u32 send_buf_index;
 	u32 total_data_buflen;
+	struct hv_dma_range *dma_range;
 };
 
 #define NETVSC_HASH_KEYLEN 40
@@ -236,6 +238,7 @@ int netvsc_recv_callback(struct net_device *net,
 void netvsc_channel_cb(void *context);
 int netvsc_poll(struct napi_struct *napi, int budget);
 
+void netvsc_xdp_xmit(struct sk_buff *skb, struct net_device *ndev);
 u32 netvsc_run_xdp(struct net_device *ndev, struct netvsc_channel *nvchan,
 		   struct xdp_buff *xdp);
 unsigned int netvsc_xdp_fraglen(unsigned int len);
@@ -245,6 +248,8 @@ int netvsc_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 		   struct netvsc_device *nvdev);
 int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog);
 int netvsc_bpf(struct net_device *dev, struct netdev_bpf *bpf);
+int netvsc_ndoxdp_xmit(struct net_device *ndev, int n,
+		       struct xdp_frame **frames, u32 flags);
 
 int rndis_set_subchannel(struct net_device *ndev,
 			 struct netvsc_device *nvdev,
@@ -941,12 +946,21 @@ struct nvsc_rsc {
 #define NVSC_RSC_CSUM_INFO	BIT(1)	/* valid/present bit for 'csum_info' */
 #define NVSC_RSC_HASH_INFO	BIT(2)	/* valid/present bit for 'hash_info' */
 
-struct netvsc_stats {
+struct netvsc_stats_tx {
+	u64 packets;
+	u64 bytes;
+	u64 xdp_xmit;
+	struct u64_stats_sync syncp;
+};
+
+struct netvsc_stats_rx {
 	u64 packets;
 	u64 bytes;
 	u64 broadcast;
 	u64 multicast;
 	u64 xdp_drop;
+	u64 xdp_redirect;
+	u64 xdp_tx;
 	struct u64_stats_sync syncp;
 };
 
@@ -1037,7 +1051,8 @@ struct net_device_context {
 	u32 vf_alloc;
 	/* Serial number of the VF to team with */
 	u32 vf_serial;
-
+	/* completion variable to confirm vf association */
+	struct completion vf_add;
 	/* Is the current data path through the VF NIC? */
 	bool  data_path_is_vf;
 
@@ -1045,6 +1060,55 @@ struct net_device_context {
 	struct netvsc_device_info *saved_netvsc_dev_info;
 };
 
+/* Azure hosts don't support non-TCP port numbers in hashing for fragmented
+ * packets. We can use ethtool to change UDP hash level when necessary.
+ */
+static inline u32 netvsc_get_hash(struct sk_buff *skb,
+				  const struct net_device_context *ndc)
+{
+	struct flow_keys flow;
+	u32 hash, pkt_proto = 0;
+	static u32 hashrnd __read_mostly;
+
+	net_get_random_once(&hashrnd, sizeof(hashrnd));
+
+	if (!skb_flow_dissect_flow_keys(skb, &flow, 0))
+		return 0;
+
+	switch (flow.basic.ip_proto) {
+	case IPPROTO_TCP:
+		if (flow.basic.n_proto == htons(ETH_P_IP))
+			pkt_proto = HV_TCP4_L4HASH;
+		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
+			pkt_proto = HV_TCP6_L4HASH;
+
+		break;
+
+	case IPPROTO_UDP:
+		if (flow.basic.n_proto == htons(ETH_P_IP))
+			pkt_proto = HV_UDP4_L4HASH;
+		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
+			pkt_proto = HV_UDP6_L4HASH;
+
+		break;
+	}
+
+	if (pkt_proto & ndc->l4_hash) {
+		return skb_get_hash(skb);
+	} else {
+		if (flow.basic.n_proto == htons(ETH_P_IP))
+			hash = jhash2((u32 *)&flow.addrs.v4addrs, 2, hashrnd);
+		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
+			hash = jhash2((u32 *)&flow.addrs.v6addrs, 8, hashrnd);
+		else
+			return 0;
+
+		__skb_set_sw_hash(skb, hash, false);
+	}
+
+	return hash;
+}
+
 /* Per channel data */
 struct netvsc_channel {
 	struct vmbus_channel *channel;
@@ -1059,9 +1123,10 @@ struct netvsc_channel {
 
 	struct bpf_prog __rcu *bpf_prog;
 	struct xdp_rxq_info xdp_rxq;
+	bool xdp_flush;
 
-	struct netvsc_stats tx_stats;
-	struct netvsc_stats rx_stats;
+	struct netvsc_stats_tx tx_stats;
+	struct netvsc_stats_rx rx_stats;
 };
 
 /* Per netvsc device */
@@ -1074,6 +1139,7 @@ struct netvsc_device {
 
 	/* Receive buffer allocated by us but manages by NetVSP */
 	void *recv_buf;
+	void *recv_original_buf;
 	u32 recv_buf_size; /* allocated bytes */
 	struct vmbus_gpadl recv_buf_gpadl_handle;
 	u32 recv_section_cnt;
@@ -1082,6 +1148,7 @@ struct netvsc_device {
 
 	/* Send buffer allocated by us */
 	void *send_buf;
+	void *send_original_buf;
 	u32 send_buf_size;
 	struct vmbus_gpadl send_buf_gpadl_handle;
 	u32 send_section_cnt;
@@ -1731,4 +1798,6 @@ struct rndis_message {
 #define RETRY_US_HI	10000
 #define RETRY_MAX	2000	/* >10 sec */
 
+void netvsc_dma_unmap(struct hv_device *hv_dev,
+		      struct hv_netvsc_packet *packet);
 #endif /* _HYPERV_NET_H */
diff --git a/drivers/net/hyperv/netvsc.c b/drivers/net/hyperv/netvsc.c
index 396bc1c204e6..9352dad58996 100644
--- a/drivers/net/hyperv/netvsc.c
+++ b/drivers/net/hyperv/netvsc.c
@@ -20,6 +20,7 @@
 #include <linux/vmalloc.h>
 #include <linux/rtnetlink.h>
 #include <linux/prefetch.h>
+#include <linux/filter.h>
 
 #include <asm/sync_bitops.h>
 #include <asm/mshyperv.h>
@@ -153,9 +154,18 @@ static void free_netvsc_device(struct rcu_head *head)
 	int i;
 
 	kfree(nvdev->extension);
-	vfree(nvdev->recv_buf);
-	vfree(nvdev->send_buf);
-	kfree(nvdev->send_section_map);
+
+	if (nvdev->recv_original_buf)
+		vfree(nvdev->recv_original_buf);
+	else
+		vfree(nvdev->recv_buf);
+
+	if (nvdev->send_original_buf)
+		vfree(nvdev->send_original_buf);
+	else
+		vfree(nvdev->send_buf);
+
+	bitmap_free(nvdev->send_section_map);
 
 	for (i = 0; i < VRSS_CHANNEL_MAX; i++) {
 		xdp_rxq_info_unreg(&nvdev->chan_table[i].xdp_rxq);
@@ -336,8 +346,8 @@ static int netvsc_init_buf(struct hv_device *device,
 	struct net_device *ndev = hv_get_drvdata(device);
 	struct nvsp_message *init_packet;
 	unsigned int buf_size;
-	size_t map_words;
 	int i, ret = 0;
+	void *vaddr;
 
 	/* Get receive buffer area. */
 	buf_size = device_info->recv_sections * device_info->recv_section_size;
@@ -373,6 +383,17 @@ static int netvsc_init_buf(struct hv_device *device,
 		goto cleanup;
 	}
 
+	if (hv_isolation_type_snp()) {
+		vaddr = hv_map_memory(net_device->recv_buf, buf_size);
+		if (!vaddr) {
+			ret = -ENOMEM;
+			goto cleanup;
+		}
+
+		net_device->recv_original_buf = net_device->recv_buf;
+		net_device->recv_buf = vaddr;
+	}
+
 	/* Notify the NetVsp of the gpadl handle */
 	init_packet = &net_device->channel_init_pkt;
 	memset(init_packet, 0, sizeof(struct nvsp_message));
@@ -476,6 +497,17 @@ static int netvsc_init_buf(struct hv_device *device,
 		goto cleanup;
 	}
 
+	if (hv_isolation_type_snp()) {
+		vaddr = hv_map_memory(net_device->send_buf, buf_size);
+		if (!vaddr) {
+			ret = -ENOMEM;
+			goto cleanup;
+		}
+
+		net_device->send_original_buf = net_device->send_buf;
+		net_device->send_buf = vaddr;
+	}
+
 	/* Notify the NetVsp of the gpadl handle */
 	init_packet = &net_device->channel_init_pkt;
 	memset(init_packet, 0, sizeof(struct nvsp_message));
@@ -528,10 +560,9 @@ static int netvsc_init_buf(struct hv_device *device,
 		   net_device->send_section_size, net_device->send_section_cnt);
 
 	/* Setup state for managing the send buffer. */
-	map_words = DIV_ROUND_UP(net_device->send_section_cnt, BITS_PER_LONG);
-
-	net_device->send_section_map = kcalloc(map_words, sizeof(ulong), GFP_KERNEL);
-	if (net_device->send_section_map == NULL) {
+	net_device->send_section_map = bitmap_zalloc(net_device->send_section_cnt,
+						     GFP_KERNEL);
+	if (!net_device->send_section_map) {
 		ret = -ENOMEM;
 		goto cleanup;
 	}
@@ -731,6 +762,12 @@ void netvsc_device_remove(struct hv_device *device)
 		netvsc_teardown_send_gpadl(device, net_device, ndev);
 	}
 
+	if (net_device->recv_original_buf)
+		hv_unmap_memory(net_device->recv_buf);
+
+	if (net_device->send_original_buf)
+		hv_unmap_memory(net_device->send_buf);
+
 	/* Release all resources */
 	free_netvsc_device_rcu(net_device);
 }
@@ -756,9 +793,9 @@ static void netvsc_send_tx_complete(struct net_device *ndev,
 	int queue_sends;
 	u64 cmd_rqst;
 
-	cmd_rqst = channel->request_addr_callback(channel, (u64)desc->trans_id);
+	cmd_rqst = channel->request_addr_callback(channel, desc->trans_id);
 	if (cmd_rqst == VMBUS_RQST_ERROR) {
-		netdev_err(ndev, "Incorrect transaction id\n");
+		netdev_err(ndev, "Invalid transaction ID %llx\n", desc->trans_id);
 		return;
 	}
 
@@ -766,10 +803,10 @@ static void netvsc_send_tx_complete(struct net_device *ndev,
 
 	/* Notify the layer above us */
 	if (likely(skb)) {
-		const struct hv_netvsc_packet *packet
+		struct hv_netvsc_packet *packet
 			= (struct hv_netvsc_packet *)skb->cb;
 		u32 send_index = packet->send_buf_index;
-		struct netvsc_stats *tx_stats;
+		struct netvsc_stats_tx *tx_stats;
 
 		if (send_index != NETVSC_INVALID_INDEX)
 			netvsc_free_send_slot(net_device, send_index);
@@ -782,6 +819,7 @@ static void netvsc_send_tx_complete(struct net_device *ndev,
 		tx_stats->bytes += packet->total_bytes;
 		u64_stats_update_end(&tx_stats->syncp);
 
+		netvsc_dma_unmap(ndev_ctx->device_ctx, packet);
 		napi_consume_skb(skb, budget);
 	}
 
@@ -817,9 +855,9 @@ static void netvsc_send_completion(struct net_device *ndev,
 	/* First check if this is a VMBUS completion without data payload */
 	if (!msglen) {
 		cmd_rqst = incoming_channel->request_addr_callback(incoming_channel,
-								   (u64)desc->trans_id);
+								   desc->trans_id);
 		if (cmd_rqst == VMBUS_RQST_ERROR) {
-			netdev_err(ndev, "Invalid transaction id\n");
+			netdev_err(ndev, "Invalid transaction ID %llx\n", desc->trans_id);
 			return;
 		}
 
@@ -946,6 +984,88 @@ static void netvsc_copy_to_send_buf(struct netvsc_device *net_device,
 		memset(dest, 0, padding);
 }
 
+void netvsc_dma_unmap(struct hv_device *hv_dev,
+		      struct hv_netvsc_packet *packet)
+{
+	u32 page_count = packet->cp_partial ?
+		packet->page_buf_cnt - packet->rmsg_pgcnt :
+		packet->page_buf_cnt;
+	int i;
+
+	if (!hv_is_isolation_supported())
+		return;
+
+	if (!packet->dma_range)
+		return;
+
+	for (i = 0; i < page_count; i++)
+		dma_unmap_single(&hv_dev->device, packet->dma_range[i].dma,
+				 packet->dma_range[i].mapping_size,
+				 DMA_TO_DEVICE);
+
+	kfree(packet->dma_range);
+}
+
+/* netvsc_dma_map - Map swiotlb bounce buffer with data page of
+ * packet sent by vmbus_sendpacket_pagebuffer() in the Isolation
+ * VM.
+ *
+ * In isolation VM, netvsc send buffer has been marked visible to
+ * host and so the data copied to send buffer doesn't need to use
+ * bounce buffer. The data pages handled by vmbus_sendpacket_pagebuffer()
+ * may not be copied to send buffer and so these pages need to be
+ * mapped with swiotlb bounce buffer. netvsc_dma_map() is to do
+ * that. The pfns in the struct hv_page_buffer need to be converted
+ * to bounce buffer's pfn. The loop here is necessary because the
+ * entries in the page buffer array are not necessarily full
+ * pages of data.  Each entry in the array has a separate offset and
+ * len that may be non-zero, even for entries in the middle of the
+ * array.  And the entries are not physically contiguous.  So each
+ * entry must be individually mapped rather than as a contiguous unit.
+ * So not use dma_map_sg() here.
+ */
+static int netvsc_dma_map(struct hv_device *hv_dev,
+			  struct hv_netvsc_packet *packet,
+			  struct hv_page_buffer *pb)
+{
+	u32 page_count =  packet->cp_partial ?
+		packet->page_buf_cnt - packet->rmsg_pgcnt :
+		packet->page_buf_cnt;
+	dma_addr_t dma;
+	int i;
+
+	if (!hv_is_isolation_supported())
+		return 0;
+
+	packet->dma_range = kcalloc(page_count,
+				    sizeof(*packet->dma_range),
+				    GFP_KERNEL);
+	if (!packet->dma_range)
+		return -ENOMEM;
+
+	for (i = 0; i < page_count; i++) {
+		char *src = phys_to_virt((pb[i].pfn << HV_HYP_PAGE_SHIFT)
+					 + pb[i].offset);
+		u32 len = pb[i].len;
+
+		dma = dma_map_single(&hv_dev->device, src, len,
+				     DMA_TO_DEVICE);
+		if (dma_mapping_error(&hv_dev->device, dma)) {
+			kfree(packet->dma_range);
+			return -ENOMEM;
+		}
+
+		/* pb[].offset and pb[].len are not changed during dma mapping
+		 * and so not reassign.
+		 */
+		packet->dma_range[i].dma = dma;
+		packet->dma_range[i].mapping_size = len;
+		pb[i].pfn = dma >> HV_HYP_PAGE_SHIFT;
+	}
+
+	return 0;
+}
+
 static inline int netvsc_send_pkt(
 	struct hv_device *device,
 	struct hv_netvsc_packet *packet,
@@ -986,14 +1106,24 @@ static inline int netvsc_send_pkt(
 
 	trace_nvsp_send_pkt(ndev, out_channel, rpkt);
 
+	packet->dma_range = NULL;
 	if (packet->page_buf_cnt) {
 		if (packet->cp_partial)
 			pb += packet->rmsg_pgcnt;
 
+		ret = netvsc_dma_map(ndev_ctx->device_ctx, packet, pb);
+		if (ret) {
+			ret = -EAGAIN;
+			goto exit;
+		}
+
 		ret = vmbus_sendpacket_pagebuffer(out_channel,
 						  pb, packet->page_buf_cnt,
 						  &nvmsg, sizeof(nvmsg),
 						  req_id);
+
+		if (ret)
+			netvsc_dma_unmap(ndev_ctx->device_ctx, packet);
 	} else {
 		ret = vmbus_sendpacket(out_channel,
 				       &nvmsg, sizeof(nvmsg),
@@ -1001,6 +1131,7 @@ static inline int netvsc_send_pkt(
 				       VMBUS_DATA_PACKET_FLAG_COMPLETION_REQUESTED);
 	}
 
+exit:
 	if (ret == 0) {
 		atomic_inc_return(&nvchan->queue_sends);
 
@@ -1449,6 +1580,10 @@ static void netvsc_send_vf(struct net_device *ndev,
 
 	net_device_ctx->vf_alloc = nvmsg->msg.v4_msg.vf_assoc.allocated;
 	net_device_ctx->vf_serial = nvmsg->msg.v4_msg.vf_assoc.serial;
+
+	if (net_device_ctx->vf_alloc)
+		complete(&net_device_ctx->vf_add);
+
 	netdev_info(ndev, "VF slot %u %s\n",
 		    net_device_ctx->vf_serial,
 		    net_device_ctx->vf_alloc ? "added" : "removed");
@@ -1500,7 +1635,6 @@ static int netvsc_process_raw_pkt(struct hv_device *device,
 
 	case VM_PKT_DATA_USING_XFER_PAGES:
 		return netvsc_receive(ndev, net_device, nvchan, desc);
-		break;
 
 	case VM_PKT_DATA_INBAND:
 		netvsc_receive_inband(ndev, net_device, desc);
@@ -1541,12 +1675,17 @@ int netvsc_poll(struct napi_struct *napi, int budget)
 	if (!nvchan->desc)
 		nvchan->desc = hv_pkt_iter_first(channel);
 
+	nvchan->xdp_flush = false;
+
 	while (nvchan->desc && work_done < budget) {
 		work_done += netvsc_process_raw_pkt(device, nvchan, net_device,
 						    ndev, nvchan->desc, budget);
 		nvchan->desc = hv_pkt_iter_next(channel, nvchan->desc);
 	}
 
+	if (nvchan->xdp_flush)
+		xdp_do_flush();
+
 	/* Send any pending receive completions */
 	ret = send_recv_completions(ndev, net_device, nvchan);
 
@@ -1644,8 +1783,7 @@ struct netvsc_device *netvsc_device_add(struct hv_device *device,
 	}
 
 	/* Enable NAPI handler before init callbacks */
-	netif_napi_add(ndev, &net_device->chan_table[0].napi,
-		       netvsc_poll, NAPI_POLL_WEIGHT);
+	netif_napi_add(ndev, &net_device->chan_table[0].napi, netvsc_poll);
 
 	/* Open the channel */
 	device->channel->next_request_id_callback = vmbus_next_request_id;
@@ -1693,6 +1831,12 @@ cleanup:
 	netif_napi_del(&net_device->chan_table[0].napi);
 
 cleanup2:
+	if (net_device->recv_original_buf)
+		hv_unmap_memory(net_device->recv_buf);
+
+	if (net_device->send_original_buf)
+		hv_unmap_memory(net_device->send_buf);
+
 	free_netvsc_device(&net_device->rcu);
 
 	return ERR_PTR(ret);
diff --git a/drivers/net/hyperv/netvsc_bpf.c b/drivers/net/hyperv/netvsc_bpf.c
index aa877da113f8..4a9522689fa4 100644
--- a/drivers/net/hyperv/netvsc_bpf.c
+++ b/drivers/net/hyperv/netvsc_bpf.c
@@ -10,6 +10,7 @@
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
+#include <linux/netpoll.h>
 #include <linux/bpf.h>
 #include <linux/bpf_trace.h>
 #include <linux/kernel.h>
@@ -23,11 +24,13 @@
 u32 netvsc_run_xdp(struct net_device *ndev, struct netvsc_channel *nvchan,
 		   struct xdp_buff *xdp)
 {
+	struct netvsc_stats_rx *rx_stats = &nvchan->rx_stats;
 	void *data = nvchan->rsc.data[0];
 	u32 len = nvchan->rsc.len[0];
 	struct page *page = NULL;
 	struct bpf_prog *prog;
 	u32 act = XDP_PASS;
+	bool drop = true;
 
 	xdp->data_hard_start = NULL;
 
@@ -60,21 +63,46 @@ u32 netvsc_run_xdp(struct net_device *ndev, struct netvsc_channel *nvchan,
 	switch (act) {
 	case XDP_PASS:
 	case XDP_TX:
+		drop = false;
+		break;
+
 	case XDP_DROP:
 		break;
 
+	case XDP_REDIRECT:
+		if (!xdp_do_redirect(ndev, xdp, prog)) {
+			nvchan->xdp_flush = true;
+			drop = false;
+
+			u64_stats_update_begin(&rx_stats->syncp);
+
+			rx_stats->xdp_redirect++;
+			rx_stats->packets++;
+			rx_stats->bytes += nvchan->rsc.pktlen;
+
+			u64_stats_update_end(&rx_stats->syncp);
+
+			break;
+		} else {
+			u64_stats_update_begin(&rx_stats->syncp);
+			rx_stats->xdp_drop++;
+			u64_stats_update_end(&rx_stats->syncp);
+		}
+
+		fallthrough;
+
 	case XDP_ABORTED:
 		trace_xdp_exception(ndev, prog, act);
 		break;
 
 	default:
-		bpf_warn_invalid_xdp_action(act);
+		bpf_warn_invalid_xdp_action(ndev, prog, act);
 	}
 
 out:
 	rcu_read_unlock();
 
-	if (page && act != XDP_PASS && act != XDP_TX) {
+	if (page && drop) {
 		__free_page(page);
 		xdp->data_hard_start = NULL;
 	}
@@ -137,7 +165,6 @@ int netvsc_xdp_set(struct net_device *dev, struct bpf_prog *prog,
 int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog)
 {
 	struct netdev_bpf xdp;
-	bpf_op_t ndo_bpf;
 	int ret;
 
 	ASSERT_RTNL();
@@ -145,8 +172,7 @@ int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog)
 	if (!vf_netdev)
 		return 0;
 
-	ndo_bpf = vf_netdev->netdev_ops->ndo_bpf;
-	if (!ndo_bpf)
+	if (!vf_netdev->netdev_ops->ndo_bpf)
 		return 0;
 
 	memset(&xdp, 0, sizeof(xdp));
@@ -157,7 +183,7 @@ int netvsc_vf_setxdp(struct net_device *vf_netdev, struct bpf_prog *prog)
 	xdp.command = XDP_SETUP_PROG;
 	xdp.prog = prog;
 
-	ret = ndo_bpf(vf_netdev, &xdp);
+	ret = vf_netdev->netdev_ops->ndo_bpf(vf_netdev, &xdp);
 
 	if (ret && prog)
 		bpf_prog_put(prog);
@@ -199,3 +225,68 @@ int netvsc_bpf(struct net_device *dev, struct netdev_bpf *bpf)
 		return -EINVAL;
 	}
 }
+
+static int netvsc_ndoxdp_xmit_fm(struct net_device *ndev,
+				 struct xdp_frame *frame, u16 q_idx)
+{
+	struct sk_buff *skb;
+
+	skb = xdp_build_skb_from_frame(frame, ndev);
+	if (unlikely(!skb))
+		return -ENOMEM;
+
+	netvsc_get_hash(skb, netdev_priv(ndev));
+
+	skb_record_rx_queue(skb, q_idx);
+
+	netvsc_xdp_xmit(skb, ndev);
+
+	return 0;
+}
+
+int netvsc_ndoxdp_xmit(struct net_device *ndev, int n,
+		       struct xdp_frame **frames, u32 flags)
+{
+	struct net_device_context *ndev_ctx = netdev_priv(ndev);
+	const struct net_device_ops *vf_ops;
+	struct netvsc_stats_tx *tx_stats;
+	struct netvsc_device *nvsc_dev;
+	struct net_device *vf_netdev;
+	int i, count = 0;
+	u16 q_idx;
+
+	/* Don't transmit if netvsc_device is gone */
+	nvsc_dev = rcu_dereference_bh(ndev_ctx->nvdev);
+	if (unlikely(!nvsc_dev || nvsc_dev->destroy))
+		return 0;
+
+	/* If VF is present and up then redirect packets to it.
+	 * Skip the VF if it is marked down or has no carrier.
+	 * If netpoll is in uses, then VF can not be used either.
+	 */
+	vf_netdev = rcu_dereference_bh(ndev_ctx->vf_netdev);
+	if (vf_netdev && netif_running(vf_netdev) &&
+	    netif_carrier_ok(vf_netdev) && !netpoll_tx_running(ndev) &&
+	    vf_netdev->netdev_ops->ndo_xdp_xmit &&
+	    ndev_ctx->data_path_is_vf) {
+		vf_ops = vf_netdev->netdev_ops;
+		return vf_ops->ndo_xdp_xmit(vf_netdev, n, frames, flags);
+	}
+
+	q_idx = smp_processor_id() % ndev->real_num_tx_queues;
+
+	for (i = 0; i < n; i++) {
+		if (netvsc_ndoxdp_xmit_fm(ndev, frames[i], q_idx))
+			break;
+
+		count++;
+	}
+
+	tx_stats = &nvsc_dev->chan_table[q_idx].tx_stats;
+
+	u64_stats_update_begin(&tx_stats->syncp);
+	tx_stats->xdp_xmit += count;
+	u64_stats_update_end(&tx_stats->syncp);
+
+	return count;
+}
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 7e66ae1d2a59..89eb4f179a3c 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -242,56 +242,6 @@ static inline void *init_ppi_data(struct rndis_message *msg,
 	return ppi + 1;
 }
 
-/* Azure hosts don't support non-TCP port numbers in hashing for fragmented
- * packets. We can use ethtool to change UDP hash level when necessary.
- */
-static inline u32 netvsc_get_hash(
-	struct sk_buff *skb,
-	const struct net_device_context *ndc)
-{
-	struct flow_keys flow;
-	u32 hash, pkt_proto = 0;
-	static u32 hashrnd __read_mostly;
-
-	net_get_random_once(&hashrnd, sizeof(hashrnd));
-
-	if (!skb_flow_dissect_flow_keys(skb, &flow, 0))
-		return 0;
-
-	switch (flow.basic.ip_proto) {
-	case IPPROTO_TCP:
-		if (flow.basic.n_proto == htons(ETH_P_IP))
-			pkt_proto = HV_TCP4_L4HASH;
-		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
-			pkt_proto = HV_TCP6_L4HASH;
-
-		break;
-
-	case IPPROTO_UDP:
-		if (flow.basic.n_proto == htons(ETH_P_IP))
-			pkt_proto = HV_UDP4_L4HASH;
-		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
-			pkt_proto = HV_UDP6_L4HASH;
-
-		break;
-	}
-
-	if (pkt_proto & ndc->l4_hash) {
-		return skb_get_hash(skb);
-	} else {
-		if (flow.basic.n_proto == htons(ETH_P_IP))
-			hash = jhash2((u32 *)&flow.addrs.v4addrs, 2, hashrnd);
-		else if (flow.basic.n_proto == htons(ETH_P_IPV6))
-			hash = jhash2((u32 *)&flow.addrs.v6addrs, 8, hashrnd);
-		else
-			return 0;
-
-		__skb_set_sw_hash(skb, hash, false);
-	}
-
-	return hash;
-}
-
 static inline int netvsc_get_tx_queue(struct net_device *ndev,
 				      struct sk_buff *skb, int old_idx)
 {
@@ -804,7 +754,7 @@ void netvsc_linkstatus_callback(struct net_device *net,
 }
 
 /* This function should only be called after skb_record_rx_queue() */
-static void netvsc_xdp_xmit(struct sk_buff *skb, struct net_device *ndev)
+void netvsc_xdp_xmit(struct sk_buff *skb, struct net_device *ndev)
 {
 	int rc;
 
@@ -925,7 +875,7 @@ int netvsc_recv_callback(struct net_device *net,
 	struct vmbus_channel *channel = nvchan->channel;
 	u16 q_idx = channel->offermsg.offer.sub_channel_index;
 	struct sk_buff *skb;
-	struct netvsc_stats *rx_stats = &nvchan->rx_stats;
+	struct netvsc_stats_rx *rx_stats = &nvchan->rx_stats;
 	struct xdp_buff xdp;
 	u32 act;
 
@@ -934,6 +884,9 @@ int netvsc_recv_callback(struct net_device *net,
 
 	act = netvsc_run_xdp(net, nvchan, &xdp);
 
+	if (act == XDP_REDIRECT)
+		return NVSP_STAT_SUCCESS;
+
 	if (act != XDP_PASS && act != XDP_TX) {
 		u64_stats_update_begin(&rx_stats->syncp);
 		rx_stats->xdp_drop++;
@@ -958,6 +911,9 @@ int netvsc_recv_callback(struct net_device *net,
 	 * statistics will not work correctly.
 	 */
 	u64_stats_update_begin(&rx_stats->syncp);
+	if (act == XDP_TX)
+		rx_stats->xdp_tx++;
+
 	rx_stats->packets++;
 	rx_stats->bytes += nvchan->rsc.pktlen;
 
@@ -979,8 +935,8 @@ int netvsc_recv_callback(struct net_device *net,
 static void netvsc_get_drvinfo(struct net_device *net,
 			       struct ethtool_drvinfo *info)
 {
-	strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
+	strscpy(info->driver, KBUILD_MODNAME, sizeof(info->driver));
+	strscpy(info->fw_version, "N/A", sizeof(info->fw_version));
 }
 
 static void netvsc_get_channels(struct net_device *net,
@@ -1353,28 +1309,29 @@ static void netvsc_get_pcpu_stats(struct net_device *net,
 	/* fetch percpu stats of netvsc */
 	for (i = 0; i < nvdev->num_chn; i++) {
 		const struct netvsc_channel *nvchan = &nvdev->chan_table[i];
-		const struct netvsc_stats *stats;
+		const struct netvsc_stats_tx *tx_stats;
+		const struct netvsc_stats_rx *rx_stats;
 		struct netvsc_ethtool_pcpu_stats *this_tot =
 			&pcpu_tot[nvchan->channel->target_cpu];
 		u64 packets, bytes;
 		unsigned int start;
 
-		stats = &nvchan->tx_stats;
+		tx_stats = &nvchan->tx_stats;
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->syncp);
-			packets = stats->packets;
-			bytes = stats->bytes;
-		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&tx_stats->syncp);
+			packets = tx_stats->packets;
+			bytes = tx_stats->bytes;
+		} while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start));
 
 		this_tot->tx_bytes	+= bytes;
 		this_tot->tx_packets	+= packets;
 
-		stats = &nvchan->rx_stats;
+		rx_stats = &nvchan->rx_stats;
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->syncp);
-			packets = stats->packets;
-			bytes = stats->bytes;
-		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&rx_stats->syncp);
+			packets = rx_stats->packets;
+			bytes = rx_stats->bytes;
+		} while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start));
 
 		this_tot->rx_bytes	+= bytes;
 		this_tot->rx_packets	+= packets;
@@ -1406,27 +1363,28 @@ static void netvsc_get_stats64(struct net_device *net,
 
 	for (i = 0; i < nvdev->num_chn; i++) {
 		const struct netvsc_channel *nvchan = &nvdev->chan_table[i];
-		const struct netvsc_stats *stats;
+		const struct netvsc_stats_tx *tx_stats;
+		const struct netvsc_stats_rx *rx_stats;
 		u64 packets, bytes, multicast;
 		unsigned int start;
 
-		stats = &nvchan->tx_stats;
+		tx_stats = &nvchan->tx_stats;
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->syncp);
-			packets = stats->packets;
-			bytes = stats->bytes;
-		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&tx_stats->syncp);
+			packets = tx_stats->packets;
+			bytes = tx_stats->bytes;
+		} while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start));
 
 		t->tx_bytes	+= bytes;
 		t->tx_packets	+= packets;
 
-		stats = &nvchan->rx_stats;
+		rx_stats = &nvchan->rx_stats;
 		do {
-			start = u64_stats_fetch_begin_irq(&stats->syncp);
-			packets = stats->packets;
-			bytes = stats->bytes;
-			multicast = stats->multicast + stats->broadcast;
-		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&rx_stats->syncp);
+			packets = rx_stats->packets;
+			bytes = rx_stats->bytes;
+			multicast = rx_stats->multicast + rx_stats->broadcast;
+		} while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start));
 
 		t->rx_bytes	+= bytes;
 		t->rx_packets	+= packets;
@@ -1515,8 +1473,8 @@ static const struct {
 /* statistics per queue (rx/tx packets/bytes) */
 #define NETVSC_PCPU_STATS_LEN (num_present_cpus() * ARRAY_SIZE(pcpu_stats))
 
-/* 5 statistics per queue (rx/tx packets/bytes, rx xdp_drop) */
-#define NETVSC_QUEUE_STATS_LEN(dev) ((dev)->num_chn * 5)
+/* 8 statistics per queue (rx/tx packets/bytes, XDP actions) */
+#define NETVSC_QUEUE_STATS_LEN(dev) ((dev)->num_chn * 8)
 
 static int netvsc_get_sset_count(struct net_device *dev, int string_set)
 {
@@ -1543,12 +1501,16 @@ static void netvsc_get_ethtool_stats(struct net_device *dev,
 	struct net_device_context *ndc = netdev_priv(dev);
 	struct netvsc_device *nvdev = rtnl_dereference(ndc->nvdev);
 	const void *nds = &ndc->eth_stats;
-	const struct netvsc_stats *qstats;
+	const struct netvsc_stats_tx *tx_stats;
+	const struct netvsc_stats_rx *rx_stats;
 	struct netvsc_vf_pcpu_stats sum;
 	struct netvsc_ethtool_pcpu_stats *pcpu_sum;
 	unsigned int start;
 	u64 packets, bytes;
 	u64 xdp_drop;
+	u64 xdp_redirect;
+	u64 xdp_tx;
+	u64 xdp_xmit;
 	int i, j, cpu;
 
 	if (!nvdev)
@@ -1562,31 +1524,40 @@ static void netvsc_get_ethtool_stats(struct net_device *dev,
 		data[i++] = *(u64 *)((void *)&sum + vf_stats[j].offset);
 
 	for (j = 0; j < nvdev->num_chn; j++) {
-		qstats = &nvdev->chan_table[j].tx_stats;
+		tx_stats = &nvdev->chan_table[j].tx_stats;
 
 		do {
-			start = u64_stats_fetch_begin_irq(&qstats->syncp);
-			packets = qstats->packets;
-			bytes = qstats->bytes;
-		} while (u64_stats_fetch_retry_irq(&qstats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&tx_stats->syncp);
+			packets = tx_stats->packets;
+			bytes = tx_stats->bytes;
+			xdp_xmit = tx_stats->xdp_xmit;
+		} while (u64_stats_fetch_retry_irq(&tx_stats->syncp, start));
 		data[i++] = packets;
 		data[i++] = bytes;
+		data[i++] = xdp_xmit;
 
-		qstats = &nvdev->chan_table[j].rx_stats;
+		rx_stats = &nvdev->chan_table[j].rx_stats;
 		do {
-			start = u64_stats_fetch_begin_irq(&qstats->syncp);
-			packets = qstats->packets;
-			bytes = qstats->bytes;
-			xdp_drop = qstats->xdp_drop;
-		} while (u64_stats_fetch_retry_irq(&qstats->syncp, start));
+			start = u64_stats_fetch_begin_irq(&rx_stats->syncp);
+			packets = rx_stats->packets;
+			bytes = rx_stats->bytes;
+			xdp_drop = rx_stats->xdp_drop;
+			xdp_redirect = rx_stats->xdp_redirect;
+			xdp_tx = rx_stats->xdp_tx;
+		} while (u64_stats_fetch_retry_irq(&rx_stats->syncp, start));
 		data[i++] = packets;
 		data[i++] = bytes;
 		data[i++] = xdp_drop;
+		data[i++] = xdp_redirect;
+		data[i++] = xdp_tx;
 	}
 
 	pcpu_sum = kvmalloc_array(num_possible_cpus(),
 				  sizeof(struct netvsc_ethtool_pcpu_stats),
 				  GFP_KERNEL);
+	if (!pcpu_sum)
+		return;
+
 	netvsc_get_pcpu_stats(dev, pcpu_sum);
 	for_each_present_cpu(cpu) {
 		struct netvsc_ethtool_pcpu_stats *this_sum = &pcpu_sum[cpu];
@@ -1619,9 +1590,12 @@ static void netvsc_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 		for (i = 0; i < nvdev->num_chn; i++) {
 			ethtool_sprintf(&p, "tx_queue_%u_packets", i);
 			ethtool_sprintf(&p, "tx_queue_%u_bytes", i);
+			ethtool_sprintf(&p, "tx_queue_%u_xdp_xmit", i);
 			ethtool_sprintf(&p, "rx_queue_%u_packets", i);
 			ethtool_sprintf(&p, "rx_queue_%u_bytes", i);
 			ethtool_sprintf(&p, "rx_queue_%u_xdp_drop", i);
+			ethtool_sprintf(&p, "rx_queue_%u_xdp_redirect", i);
+			ethtool_sprintf(&p, "rx_queue_%u_xdp_tx", i);
 		}
 
 		for_each_present_cpu(cpu) {
@@ -1858,7 +1832,9 @@ static void __netvsc_get_ringparam(struct netvsc_device *nvdev,
 }
 
 static void netvsc_get_ringparam(struct net_device *ndev,
-				 struct ethtool_ringparam *ring)
+				 struct ethtool_ringparam *ring,
+				 struct kernel_ethtool_ringparam *kernel_ring,
+				 struct netlink_ext_ack *extack)
 {
 	struct net_device_context *ndevctx = netdev_priv(ndev);
 	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
@@ -1870,7 +1846,9 @@ static void netvsc_get_ringparam(struct net_device *ndev,
 }
 
 static int netvsc_set_ringparam(struct net_device *ndev,
-				struct ethtool_ringparam *ring)
+				struct ethtool_ringparam *ring,
+				struct kernel_ethtool_ringparam *kernel_ring,
+				struct netlink_ext_ack *extack)
 {
 	struct net_device_context *ndevctx = netdev_priv(ndev);
 	struct netvsc_device *nvdev = rtnl_dereference(ndevctx->nvdev);
@@ -2050,6 +2028,7 @@ static const struct net_device_ops device_ops = {
 	.ndo_select_queue =		netvsc_select_queue,
 	.ndo_get_stats64 =		netvsc_get_stats64,
 	.ndo_bpf =			netvsc_bpf,
+	.ndo_xdp_xmit =			netvsc_ndoxdp_xmit,
 };
 
 /*
@@ -2334,6 +2313,18 @@ static struct net_device *get_netvsc_byslot(const struct net_device *vf_netdev)
 
 	}
 
+	/* Fallback path to check synthetic vf with
+	 * help of mac addr
+	 */
+	list_for_each_entry(ndev_ctx, &netvsc_dev_list, list) {
+		ndev = hv_get_drvdata(ndev_ctx->device_ctx);
+		if (ether_addr_equal(vf_netdev->perm_addr, ndev->perm_addr)) {
+			netdev_notice(vf_netdev,
+				      "falling back to mac addr based matching\n");
+			return ndev;
+		}
+	}
+
 	netdev_notice(vf_netdev,
 		      "no netdev found for vf serial:%u\n", serial);
 	return NULL;
@@ -2430,6 +2421,11 @@ static int netvsc_vf_changed(struct net_device *vf_netdev, unsigned long event)
 	if (net_device_ctx->data_path_is_vf == vf_is_up)
 		return NOTIFY_OK;
 
+	if (vf_is_up && !net_device_ctx->vf_alloc) {
+		netdev_info(ndev, "Waiting for the VF association from host\n");
+		wait_for_completion(&net_device_ctx->vf_add);
+	}
+
 	ret = netvsc_switch_datapath(ndev, vf_is_up);
 
 	if (ret) {
@@ -2461,6 +2457,7 @@ static int netvsc_unregister_vf(struct net_device *vf_netdev)
 
 	netvsc_vf_setxdp(vf_netdev, NULL);
 
+	reinit_completion(&net_device_ctx->vf_add);
 	netdev_rx_handler_unregister(vf_netdev);
 	netdev_upper_dev_unlink(vf_netdev, ndev);
 	RCU_INIT_POINTER(net_device_ctx->vf_netdev, NULL);
@@ -2500,6 +2497,7 @@ static int netvsc_probe(struct hv_device *dev,
 
 	INIT_DELAYED_WORK(&net_device_ctx->dwork, netvsc_link_change);
 
+	init_completion(&net_device_ctx->vf_add);
 	spin_lock_init(&net_device_ctx->lock);
 	INIT_LIST_HEAD(&net_device_ctx->reconfig_events);
 	INIT_DELAYED_WORK(&net_device_ctx->vf_takeover, netvsc_vf_setup);
@@ -2512,6 +2510,7 @@ static int netvsc_probe(struct hv_device *dev,
 	net->netdev_ops = &device_ops;
 	net->ethtool_ops = &ethtool_ops;
 	SET_NETDEV_DEV(net, &dev->device);
+	dma_set_min_align_mask(&dev->device, HV_HYP_PAGE_SIZE - 1);
 
 	/* We always need headroom for rndis header */
 	net->needed_headroom = RNDIS_AND_PPI_SIZE;
@@ -2663,7 +2662,10 @@ static int netvsc_suspend(struct hv_device *dev)
 
 	/* Save the current config info */
 	ndev_ctx->saved_netvsc_dev_info = netvsc_devinfo_get(nvdev);
-
+	if (!ndev_ctx->saved_netvsc_dev_info) {
+		ret = -ENOMEM;
+		goto out;
+	}
 	ret = netvsc_detach(net, nvdev);
 out:
 	rtnl_unlock();
diff --git a/drivers/net/hyperv/rndis_filter.c b/drivers/net/hyperv/rndis_filter.c
index f6c9c2a670f9..eea777ec2541 100644
--- a/drivers/net/hyperv/rndis_filter.c
+++ b/drivers/net/hyperv/rndis_filter.c
@@ -20,6 +20,7 @@
 #include <linux/vmalloc.h>
 #include <linux/rtnetlink.h>
 #include <linux/ucs2_string.h>
+#include <linux/string.h>
 
 #include "hyperv_net.h"
 #include "netvsc_trace.h"
@@ -335,9 +336,10 @@ static void rndis_filter_receive_response(struct net_device *ndev,
 		if (resp->msg_len <=
 		    sizeof(struct rndis_message) + RNDIS_EXT_LEN) {
 			memcpy(&request->response_msg, resp, RNDIS_HEADER_SIZE + sizeof(*req_id));
-			memcpy((void *)&request->response_msg + RNDIS_HEADER_SIZE + sizeof(*req_id),
+			unsafe_memcpy((void *)&request->response_msg + RNDIS_HEADER_SIZE + sizeof(*req_id),
 			       data + RNDIS_HEADER_SIZE + sizeof(*req_id),
-			       resp->msg_len - RNDIS_HEADER_SIZE - sizeof(*req_id));
+			       resp->msg_len - RNDIS_HEADER_SIZE - sizeof(*req_id),
+			       "request->response_msg is followed by a padding of RNDIS_EXT_LEN inside rndis_request");
 			if (request->request_msg.ndis_msg_type ==
 			    RNDIS_MSG_QUERY && request->request_msg.msg.
 			    query_req.oid == RNDIS_OID_GEN_MEDIA_CONNECT_STATUS)
@@ -361,6 +363,8 @@ static void rndis_filter_receive_response(struct net_device *ndev,
 			}
 		}
 
+		netvsc_dma_unmap(((struct net_device_context *)
+			netdev_priv(ndev))->device_ctx, &request->pkt);
 		complete(&request->wait_event);
 	} else {
 		netdev_err(ndev,
@@ -1347,7 +1351,7 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	struct net_device_context *net_device_ctx = netdev_priv(net);
 	struct ndis_offload hwcaps;
 	struct ndis_offload_params offloads;
-	unsigned int gso_max_size = GSO_MAX_SIZE;
+	unsigned int gso_max_size = GSO_LEGACY_MAX_SIZE;
 	int ret;
 
 	/* Find HW offload capabilities */
@@ -1429,7 +1433,7 @@ static int rndis_netdev_set_hwcaps(struct rndis_device *rndis_device,
 	 */
 	net->features &= ~NETVSC_SUPPORTED_HW_FEATURES | net->hw_features;
 
-	netif_set_gso_max_size(net, gso_max_size);
+	netif_set_tso_max_size(net, gso_max_size);
 
 	ret = rndis_filter_set_offload_params(net, nvdev, &offloads);
 
@@ -1573,7 +1577,7 @@ struct netvsc_device *rndis_filter_device_add(struct hv_device *dev,
 
 	for (i = 1; i < net_device->num_chn; i++)
 		netif_napi_add(net, &net_device->chan_table[i].napi,
-			       netvsc_poll, NAPI_POLL_WEIGHT);
+			       netvsc_poll);
 
 	return net_device;