Merge ~paelzer/ubuntu/+source/dpdk:lp-1940913-MRE-20.11.3-IMPISH into ubuntu/+source/dpdk:ubuntu/impish-devel

[Christian Ehrhardt ]

[Christian Ehrhardt  (paelzer)]

PPA: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/4666/+packages

[Christian Ehrhardt  (paelzer)]

FYI this follows Debians model of importing the upstream not as a single commit, but as

To ease comparison I recommend from <email address hidden>:debian/dpdk.git the branch upstream-20.11-stable

With that you can easily see that it matches the upstream release and other than that doesn't change too much.

[Bryce Harrington (bryce)]

LGTM, see focal MP for detail.

review: Approve

[Christian Ehrhardt  (paelzer)]

Thank you, uploaded

Uploading to ubuntu (via ftp to upload.ubuntu.com):
  Uploading dpdk_20.11.3-0ubuntu1.dsc: done.
  Uploading dpdk_20.11.3.orig.tar.xz: done.
  Uploading dpdk_20.11.3-0ubuntu1.debian.tar.xz: done.
  Uploading dpdk_20.11.3-0ubuntu1_source.buildinfo: done.
  Uploading dpdk_20.11.3-0ubuntu1_source.changes: done.
Successfully uploaded packages.

diff --git a/MAINTAINERS b/MAINTAINERS
index f45c8c1..dcde2ab 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -103,6 +103,7 @@ F: buildtools/gen-pmdinfo-cfile.sh
 F: buildtools/list-dir-globs.py
 F: buildtools/pkg-config/
 F: buildtools/symlink-drivers-solibs.sh
+F: buildtools/symlink-drivers-solibs.py
 F: devtools/test-meson-builds.sh
 
 Public CI
diff --git a/VERSION b/VERSION
index 0951ac2..bc6fac4 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-20.11.2
+20.11.3
diff --git a/app/test-crypto-perf/cperf_test_common.c b/app/test-crypto-perf/cperf_test_common.c
index 058e0ba..12925c7 100644
--- a/app/test-crypto-perf/cperf_test_common.c
+++ b/app/test-crypto-perf/cperf_test_common.c
@@ -194,7 +194,7 @@ cperf_alloc_common_memory(const struct cperf_options *options,
 				(mbuf_size * segments_nb);
 		params.dst_buf_offset = *dst_buf_offset;
 		/* Destination buffer will be one segment only */
-		obj_size += max_size;
+		obj_size += max_size + sizeof(struct rte_mbuf);
 	}
 
 	*pool = rte_mempool_create_empty(pool_name,
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 3712daa..58ebff5 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -1225,7 +1225,7 @@ cmdline_parse_token_string_t cmd_operate_port_all_all =
 cmdline_parse_inst_t cmd_operate_port = {
 	.f = cmd_operate_port_parsed,
 	.data = NULL,
-	.help_str = "port start|stop|close all: Start/Stop/Close/Reset all ports",
+	.help_str = "port start|stop|close|reset all: Start/Stop/Close/Reset all ports",
 	.tokens = {
 		(void *)&cmd_operate_port_all_cmd,
 		(void *)&cmd_operate_port_all_port,
@@ -1272,7 +1272,7 @@ cmdline_parse_token_num_t cmd_operate_specific_port_id =
 cmdline_parse_inst_t cmd_operate_specific_port = {
 	.f = cmd_operate_specific_port_parsed,
 	.data = NULL,
-	.help_str = "port start|stop|close <port_id>: Start/Stop/Close/Reset port_id",
+	.help_str = "port start|stop|close|reset <port_id>: Start/Stop/Close/Reset port_id",
 	.tokens = {
 		(void *)&cmd_operate_specific_port_cmd,
 		(void *)&cmd_operate_specific_port_port,
@@ -1607,13 +1607,13 @@ cmd_config_speed_specific_parsed(void *parsed_result,
 	struct cmd_config_speed_specific *res = parsed_result;
 	uint32_t link_speed;
 
-	if (!all_ports_stopped()) {
-		printf("Please stop all ports first\n");
+	if (port_id_is_invalid(res->id, ENABLED_WARN))
 		return;
-	}
 
-	if (port_id_is_invalid(res->id, ENABLED_WARN))
+	if (!port_is_stopped(res->id)) {
+		printf("Please stop port %d first\n", res->id);
 		return;
+	}
 
 	if (parse_and_check_speed_duplex(res->value1, res->value2,
 			&link_speed) < 0)
@@ -16449,17 +16449,17 @@ cmd_set_port_fec_mode_parsed(
 {
 	struct cmd_set_port_fec_mode *res = parsed_result;
 	uint16_t port_id = res->port_id;
-	uint32_t mode;
+	uint32_t fec_capa;
 	int ret;
 
-	ret = parse_fec_mode(res->fec_value, &mode);
+	ret = parse_fec_mode(res->fec_value, &fec_capa);
 	if (ret < 0) {
 		printf("Unknown fec mode: %s for Port %d\n", res->fec_value,
 			port_id);
 		return;
 	}
 
-	ret = rte_eth_fec_set(port_id, mode);
+	ret = rte_eth_fec_set(port_id, fec_capa);
 	if (ret == -ENOTSUP) {
 		printf("Function not implemented\n");
 		return;
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 937a90f..4847c36 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -3760,13 +3760,14 @@ set_tx_pkt_split(const char *name)
 }
 
 int
-parse_fec_mode(const char *name, uint32_t *mode)
+parse_fec_mode(const char *name, uint32_t *fec_capa)
 {
 	uint8_t i;
 
 	for (i = 0; i < RTE_DIM(fec_mode_name); i++) {
 		if (strcmp(fec_mode_name[i].name, name) == 0) {
-			*mode = RTE_ETH_FEC_MODE_TO_CAPA(fec_mode_name[i].mode);
+			*fec_capa =
+				RTE_ETH_FEC_MODE_TO_CAPA(fec_mode_name[i].mode);
 			return 0;
 		}
 	}
diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c
index d813d4f..b0a58e8 100644
--- a/app/test-pmd/csumonly.c
+++ b/app/test-pmd/csumonly.c
@@ -480,17 +480,18 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
 
 	if (info->ethertype == _htons(RTE_ETHER_TYPE_IPV4)) {
 		ipv4_hdr = l3_hdr;
-		ipv4_hdr->hdr_checksum = 0;
 
 		ol_flags |= PKT_TX_IPV4;
 		if (info->l4_proto == IPPROTO_TCP && tso_segsz) {
 			ol_flags |= PKT_TX_IP_CKSUM;
 		} else {
-			if (tx_offloads & DEV_TX_OFFLOAD_IPV4_CKSUM)
+			if (tx_offloads & DEV_TX_OFFLOAD_IPV4_CKSUM) {
 				ol_flags |= PKT_TX_IP_CKSUM;
-			else
+			} else {
+				ipv4_hdr->hdr_checksum = 0;
 				ipv4_hdr->hdr_checksum =
 					rte_ipv4_cksum(ipv4_hdr);
+			}
 		}
 	} else if (info->ethertype == _htons(RTE_ETHER_TYPE_IPV6))
 		ol_flags |= PKT_TX_IPV6;
@@ -501,10 +502,10 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
 		udp_hdr = (struct rte_udp_hdr *)((char *)l3_hdr + info->l3_len);
 		/* do not recalculate udp cksum if it was 0 */
 		if (udp_hdr->dgram_cksum != 0) {
-			udp_hdr->dgram_cksum = 0;
-			if (tx_offloads & DEV_TX_OFFLOAD_UDP_CKSUM)
+			if (tx_offloads & DEV_TX_OFFLOAD_UDP_CKSUM) {
 				ol_flags |= PKT_TX_UDP_CKSUM;
-			else {
+			} else {
+				udp_hdr->dgram_cksum = 0;
 				udp_hdr->dgram_cksum =
 					get_udptcp_checksum(l3_hdr, udp_hdr,
 						info->ethertype);
@@ -514,12 +515,12 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
 			ol_flags |= PKT_TX_UDP_SEG;
 	} else if (info->l4_proto == IPPROTO_TCP) {
 		tcp_hdr = (struct rte_tcp_hdr *)((char *)l3_hdr + info->l3_len);
-		tcp_hdr->cksum = 0;
 		if (tso_segsz)
 			ol_flags |= PKT_TX_TCP_SEG;
-		else if (tx_offloads & DEV_TX_OFFLOAD_TCP_CKSUM)
+		else if (tx_offloads & DEV_TX_OFFLOAD_TCP_CKSUM) {
 			ol_flags |= PKT_TX_TCP_CKSUM;
-		else {
+		} else {
+			tcp_hdr->cksum = 0;
 			tcp_hdr->cksum =
 				get_udptcp_checksum(l3_hdr, tcp_hdr,
 					info->ethertype);
@@ -529,13 +530,13 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
 	} else if (info->l4_proto == IPPROTO_SCTP) {
 		sctp_hdr = (struct rte_sctp_hdr *)
 			((char *)l3_hdr + info->l3_len);
-		sctp_hdr->cksum = 0;
 		/* sctp payload must be a multiple of 4 to be
 		 * offloaded */
 		if ((tx_offloads & DEV_TX_OFFLOAD_SCTP_CKSUM) &&
 			((ipv4_hdr->total_length & 0x3) == 0)) {
 			ol_flags |= PKT_TX_SCTP_CKSUM;
 		} else {
+			sctp_hdr->cksum = 0;
 			/* XXX implement CRC32c, example available in
 			 * RFC3309 */
 		}
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index d830fe3..5288c04 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -1399,22 +1399,69 @@ check_nb_hairpinq(queueid_t hairpinq)
 }
 
 static void
+init_config_port_offloads(portid_t pid, uint32_t socket_id)
+{
+	struct rte_port *port = &ports[pid];
+	uint16_t data_size;
+	int ret;
+	int i;
+
+	port->dev_conf.txmode = tx_mode;
+	port->dev_conf.rxmode = rx_mode;
+
+	ret = eth_dev_info_get_print_err(pid, &port->dev_info);
+	if (ret != 0)
+		rte_exit(EXIT_FAILURE, "rte_eth_dev_info_get() failed\n");
+
+	ret = update_jumbo_frame_offload(pid);
+	if (ret != 0)
+		printf("Updating jumbo frame offload failed for port %u\n",
+			pid);
+
+	if (!(port->dev_info.tx_offload_capa & DEV_TX_OFFLOAD_MBUF_FAST_FREE))
+		port->dev_conf.txmode.offloads &=
+			~DEV_TX_OFFLOAD_MBUF_FAST_FREE;
+
+	/* Apply Rx offloads configuration */
+	for (i = 0; i < port->dev_info.max_rx_queues; i++)
+		port->rx_conf[i].offloads = port->dev_conf.rxmode.offloads;
+	/* Apply Tx offloads configuration */
+	for (i = 0; i < port->dev_info.max_tx_queues; i++)
+		port->tx_conf[i].offloads = port->dev_conf.txmode.offloads;
+
+	/* set flag to initialize port/queue */
+	port->need_reconfig = 1;
+	port->need_reconfig_queues = 1;
+	port->socket_id = socket_id;
+	port->tx_metadata = 0;
+
+	/*
+	 * Check for maximum number of segments per MTU.
+	 * Accordingly update the mbuf data size.
+	 */
+	if (port->dev_info.rx_desc_lim.nb_mtu_seg_max != UINT16_MAX &&
+	    port->dev_info.rx_desc_lim.nb_mtu_seg_max != 0) {
+		data_size = rx_mode.max_rx_pkt_len /
+			port->dev_info.rx_desc_lim.nb_mtu_seg_max;
+
+		if ((data_size + RTE_PKTMBUF_HEADROOM) > mbuf_data_size[0]) {
+			mbuf_data_size[0] = data_size + RTE_PKTMBUF_HEADROOM;
+			TESTPMD_LOG(WARNING,
+				    "Configured mbuf size of the first segment %hu\n",
+				    mbuf_data_size[0]);
+		}
+	}
+}
+
+static void
 init_config(void)
 {
 	portid_t pid;
-	struct rte_port *port;
 	struct rte_mempool *mbp;
 	unsigned int nb_mbuf_per_pool;
 	lcoreid_t  lc_id;
-	uint8_t port_per_socket[RTE_MAX_NUMA_NODES];
 	struct rte_gro_param gro_param;
 	uint32_t gso_types;
-	uint16_t data_size;
-	bool warning = 0;
-	int k;
-	int ret;
-
-	memset(port_per_socket,0,RTE_MAX_NUMA_NODES);
 
 	/* Configuration of logical cores. */
 	fwd_lcores = rte_zmalloc("testpmd: fwd_lcores",
@@ -1436,30 +1483,12 @@ init_config(void)
 	}
 
 	RTE_ETH_FOREACH_DEV(pid) {
-		port = &ports[pid];
-		/* Apply default TxRx configuration for all ports */
-		port->dev_conf.txmode = tx_mode;
-		port->dev_conf.rxmode = rx_mode;
+		uint32_t socket_id;
 
-		ret = eth_dev_info_get_print_err(pid, &port->dev_info);
-		if (ret != 0)
-			rte_exit(EXIT_FAILURE,
-				 "rte_eth_dev_info_get() failed\n");
-
-		ret = update_jumbo_frame_offload(pid);
-		if (ret != 0)
-			printf("Updating jumbo frame offload failed for port %u\n",
-				pid);
-
-		if (!(port->dev_info.tx_offload_capa &
-		      DEV_TX_OFFLOAD_MBUF_FAST_FREE))
-			port->dev_conf.txmode.offloads &=
-				~DEV_TX_OFFLOAD_MBUF_FAST_FREE;
 		if (numa_support) {
-			if (port_numa[pid] != NUMA_NO_CONFIG)
-				port_per_socket[port_numa[pid]]++;
-			else {
-				uint32_t socket_id = rte_eth_dev_socket_id(pid);
+			socket_id = port_numa[pid];
+			if (port_numa[pid] == NUMA_NO_CONFIG) {
+				socket_id = rte_eth_dev_socket_id(pid);
 
 				/*
 				 * if socket_id is invalid,
@@ -1467,45 +1496,14 @@ init_config(void)
 				 */
 				if (check_socket_id(socket_id) < 0)
 					socket_id = socket_ids[0];
-				port_per_socket[socket_id]++;
-			}
-		}
-
-		/* Apply Rx offloads configuration */
-		for (k = 0; k < port->dev_info.max_rx_queues; k++)
-			port->rx_conf[k].offloads =
-				port->dev_conf.rxmode.offloads;
-		/* Apply Tx offloads configuration */
-		for (k = 0; k < port->dev_info.max_tx_queues; k++)
-			port->tx_conf[k].offloads =
-				port->dev_conf.txmode.offloads;
-
-		/* set flag to initialize port/queue */
-		port->need_reconfig = 1;
-		port->need_reconfig_queues = 1;
-		port->tx_metadata = 0;
-
-		/* Check for maximum number of segments per MTU. Accordingly
-		 * update the mbuf data size.
-		 */
-		if (port->dev_info.rx_desc_lim.nb_mtu_seg_max != UINT16_MAX &&
-				port->dev_info.rx_desc_lim.nb_mtu_seg_max != 0) {
-			data_size = rx_mode.max_rx_pkt_len /
-				port->dev_info.rx_desc_lim.nb_mtu_seg_max;
-
-			if ((data_size + RTE_PKTMBUF_HEADROOM) >
-							mbuf_data_size[0]) {
-				mbuf_data_size[0] = data_size +
-						 RTE_PKTMBUF_HEADROOM;
-				warning = 1;
 			}
+		} else {
+			socket_id = (socket_num == UMA_NO_CONFIG) ?
+				    0 : socket_num;
 		}
+		/* Apply default TxRx configuration for all ports */
+		init_config_port_offloads(pid, socket_id);
 	}
-
-	if (warning)
-		TESTPMD_LOG(WARNING,
-			    "Configured mbuf size of the first segment %hu\n",
-			    mbuf_data_size[0]);
 	/*
 	 * Create pools of mbuf.
 	 * If NUMA support is disabled, create a single pool of mbuf in
@@ -1592,21 +1590,8 @@ init_config(void)
 void
 reconfig(portid_t new_port_id, unsigned socket_id)
 {
-	struct rte_port *port;
-	int ret;
-
 	/* Reconfiguration of Ethernet ports. */
-	port = &ports[new_port_id];
-
-	ret = eth_dev_info_get_print_err(new_port_id, &port->dev_info);
-	if (ret != 0)
-		return;
-
-	/* set flag to initialize port/queue */
-	port->need_reconfig = 1;
-	port->need_reconfig_queues = 1;
-	port->socket_id = socket_id;
-
+	init_config_port_offloads(new_port_id, socket_id);
 	init_port_config();
 }
 
@@ -2429,7 +2414,6 @@ start_port(portid_t pid)
 	int peer_pi;
 	queueid_t qi;
 	struct rte_port *port;
-	struct rte_ether_addr mac_addr;
 	struct rte_eth_hairpin_cap cap;
 
 	if (port_id_is_invalid(pid, ENABLED_WARN))
@@ -2598,11 +2582,14 @@ start_port(portid_t pid)
 			RTE_PORT_HANDLING, RTE_PORT_STARTED) == 0)
 			printf("Port %d can not be set into started\n", pi);
 
-		if (eth_macaddr_get_print_err(pi, &mac_addr) == 0)
+		if (eth_macaddr_get_print_err(pi, &port->eth_addr) == 0)
 			printf("Port %d: %02X:%02X:%02X:%02X:%02X:%02X\n", pi,
-				mac_addr.addr_bytes[0], mac_addr.addr_bytes[1],
-				mac_addr.addr_bytes[2], mac_addr.addr_bytes[3],
-				mac_addr.addr_bytes[4], mac_addr.addr_bytes[5]);
+				port->eth_addr.addr_bytes[0],
+				port->eth_addr.addr_bytes[1],
+				port->eth_addr.addr_bytes[2],
+				port->eth_addr.addr_bytes[3],
+				port->eth_addr.addr_bytes[4],
+				port->eth_addr.addr_bytes[5]);
 
 		/* at least one port started, need checking link status */
 		need_check_link_status = 1;
diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index aad2eb3..927ca58 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h
@@ -878,7 +878,7 @@ void show_tx_pkt_segments(void);
 void set_tx_pkt_times(unsigned int *tx_times);
 void show_tx_pkt_times(void);
 void set_tx_pkt_split(const char *name);
-int parse_fec_mode(const char *name, enum rte_eth_fec_mode *mode);
+int parse_fec_mode(const char *name, uint32_t *fec_capa);
 void show_fec_capability(uint32_t num, struct rte_eth_fec_capa *speed_fec_capa);
 void set_nb_pkt_per_burst(uint16_t pkt_burst);
 char *list_pkt_forwarding_modes(void);
diff --git a/app/test/packet_burst_generator.c b/app/test/packet_burst_generator.c
index c05ea7a..0fd7290 100644
--- a/app/test/packet_burst_generator.c
+++ b/app/test/packet_burst_generator.c
@@ -142,8 +142,8 @@ uint16_t
 initialize_ipv6_header(struct rte_ipv6_hdr *ip_hdr, uint8_t *src_addr,
 		uint8_t *dst_addr, uint16_t pkt_data_len)
 {
-	ip_hdr->vtc_flow = 0;
-	ip_hdr->payload_len = pkt_data_len;
+	ip_hdr->vtc_flow = rte_cpu_to_be_32(0x60000000); /* Set version to 6. */
+	ip_hdr->payload_len = rte_cpu_to_be_16(pkt_data_len);
 	ip_hdr->proto = IPPROTO_UDP;
 	ip_hdr->hop_limits = IP_DEFTTL;
 
diff --git a/app/test/test_cmdline_lib.c b/app/test/test_cmdline_lib.c
index bd72df0..d5a09b4 100644
--- a/app/test/test_cmdline_lib.c
+++ b/app/test/test_cmdline_lib.c
@@ -71,10 +71,12 @@ test_cmdline_parse_fns(void)
 	if (cmdline_complete(cl, "buffer", &i, NULL, sizeof(dst)) >= 0)
 		goto error;
 
+	cmdline_free(cl);
 	return 0;
 
 error:
 	printf("Error: function accepted null parameter!\n");
+	cmdline_free(cl);
 	return -1;
 }
 
@@ -140,22 +142,31 @@ static int
 test_cmdline_socket_fns(void)
 {
 	cmdline_parse_ctx_t ctx;
+	struct cmdline *cl;
 
-	if (cmdline_stdin_new(NULL, "prompt") != NULL)
+	cl = cmdline_stdin_new(NULL, "prompt");
+	if (cl != NULL)
 		goto error;
-	if (cmdline_stdin_new(&ctx, NULL) != NULL)
+	cl = cmdline_stdin_new(&ctx, NULL);
+	if (cl != NULL)
 		goto error;
-	if (cmdline_file_new(NULL, "prompt", "/dev/null") != NULL)
+	cl = cmdline_file_new(NULL, "prompt", "/dev/null");
+	if (cl != NULL)
 		goto error;
-	if (cmdline_file_new(&ctx, NULL, "/dev/null") != NULL)
+	cl = cmdline_file_new(&ctx, NULL, "/dev/null");
+	if (cl != NULL)
 		goto error;
-	if (cmdline_file_new(&ctx, "prompt", NULL) != NULL)
+	cl = cmdline_file_new(&ctx, "prompt", NULL);
+	if (cl != NULL)
 		goto error;
-	if (cmdline_file_new(&ctx, "prompt", "-/invalid/~/path") != NULL) {
+	cl = cmdline_file_new(&ctx, "prompt", "-/invalid/~/path");
+	if (cl != NULL) {
 		printf("Error: succeeded in opening invalid file for reading!");
+		cmdline_free(cl);
 		return -1;
 	}
-	if (cmdline_file_new(&ctx, "prompt", "/dev/null") == NULL) {
+	cl = cmdline_file_new(&ctx, "prompt", "/dev/null");
+	if (cl == NULL) {
 		printf("Error: failed to open /dev/null for reading!");
 		return -1;
 	}
@@ -163,9 +174,11 @@ test_cmdline_socket_fns(void)
 	/* void functions */
 	cmdline_stdin_exit(NULL);
 
+	cmdline_free(cl);
 	return 0;
 error:
 	printf("Error: function accepted null parameter!\n");
+	cmdline_free(cl);
 	return -1;
 }
 
@@ -176,13 +189,14 @@ test_cmdline_fns(void)
 	struct cmdline *cl;
 
 	memset(&ctx, 0, sizeof(ctx));
-	cl = cmdline_new(&ctx, "test", -1, -1);
-	if (cl == NULL)
+	cl = cmdline_new(NULL, "prompt", 0, 0);
+	if (cl != NULL)
 		goto error;
-
-	if (cmdline_new(NULL, "prompt", 0, 0) != NULL)
+	cl = cmdline_new(&ctx, NULL, 0, 0);
+	if (cl != NULL)
 		goto error;
-	if (cmdline_new(&ctx, NULL, 0, 0) != NULL)
+	cl = cmdline_new(&ctx, "test", -1, -1);
+	if (cl == NULL)
 		goto error;
 	if (cmdline_in(NULL, "buffer", CMDLINE_TEST_BUFSIZE) >= 0)
 		goto error;
@@ -198,6 +212,7 @@ test_cmdline_fns(void)
 	cmdline_interact(NULL);
 	cmdline_quit(NULL);
 
+	cmdline_free(cl);
 	return 0;
 
 error:
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index ea24e8e..dd2b035 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -135,10 +135,11 @@ setup_test_string(struct rte_mempool *mpool,
 	struct rte_mbuf *m = rte_pktmbuf_alloc(mpool);
 	size_t t_len = len - (blocksize ? (len % blocksize) : 0);
 
-	memset(m->buf_addr, 0, m->buf_len);
 	if (m) {
-		char *dst = rte_pktmbuf_append(m, t_len);
+		char *dst;
 
+		memset(m->buf_addr, 0, m->buf_len);
+		dst = rte_pktmbuf_append(m, t_len);
 		if (!dst) {
 			rte_pktmbuf_free(m);
 			return NULL;
@@ -12024,7 +12025,7 @@ test_authenticated_decryption_fail_when_corruption(
 }
 
 static int
-test_authenticated_encryt_with_esn(
+test_authenticated_encrypt_with_esn(
 		struct crypto_testsuite_params *ts_params,
 		struct crypto_unittest_params *ut_params,
 		const struct test_crypto_vector *reference)
@@ -12811,7 +12812,7 @@ auth_decryption_AES128CBC_HMAC_SHA1_fail_tag_corrupt(void)
 static int
 auth_encrypt_AES128CBC_HMAC_SHA1_esn_check(void)
 {
-	return test_authenticated_encryt_with_esn(
+	return test_authenticated_encrypt_with_esn(
 			&testsuite_params,
 			&unittest_params,
 			&aes128cbc_hmac_sha1_aad_test_vector);
@@ -13786,7 +13787,7 @@ static struct unit_test_suite cryptodev_ccp_testsuite  = {
 };
 
 static int
-test_cryptodev_qat(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_qat(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_QAT_SYM_PMD));
@@ -13800,7 +13801,7 @@ test_cryptodev_qat(void /*argv __rte_unused, int argc __rte_unused*/)
 }
 
 static int
-test_cryptodev_virtio(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_virtio(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_VIRTIO_PMD));
@@ -13814,7 +13815,7 @@ test_cryptodev_virtio(void /*argv __rte_unused, int argc __rte_unused*/)
 }
 
 static int
-test_cryptodev_aesni_mb(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_aesni_mb(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_AESNI_MB_PMD));
@@ -13912,7 +13913,7 @@ test_cryptodev_null(void)
 }
 
 static int
-test_cryptodev_sw_snow3g(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_sw_snow3g(void)
 {
 	gbl_driver_id = rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_SNOW3G_PMD));
@@ -13926,7 +13927,7 @@ test_cryptodev_sw_snow3g(void /*argv __rte_unused, int argc __rte_unused*/)
 }
 
 static int
-test_cryptodev_sw_kasumi(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_sw_kasumi(void)
 {
 	gbl_driver_id = rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_KASUMI_PMD));
@@ -13940,7 +13941,7 @@ test_cryptodev_sw_kasumi(void /*argv __rte_unused, int argc __rte_unused*/)
 }
 
 static int
-test_cryptodev_sw_zuc(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_sw_zuc(void)
 {
 	gbl_driver_id = rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_ZUC_PMD));
@@ -13984,7 +13985,7 @@ test_cryptodev_mrvl(void)
 #ifdef RTE_CRYPTO_SCHEDULER
 
 static int
-test_cryptodev_scheduler(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_scheduler(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_SCHEDULER_PMD));
@@ -14007,7 +14008,7 @@ REGISTER_TEST_COMMAND(cryptodev_scheduler_autotest, test_cryptodev_scheduler);
 #endif
 
 static int
-test_cryptodev_dpaa2_sec(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_dpaa2_sec(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_DPAA2_SEC_PMD));
@@ -14021,7 +14022,7 @@ test_cryptodev_dpaa2_sec(void /*argv __rte_unused, int argc __rte_unused*/)
 }
 
 static int
-test_cryptodev_dpaa_sec(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_dpaa_sec(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_DPAA_SEC_PMD));
@@ -14073,7 +14074,7 @@ test_cryptodev_octeontx2(void)
 }
 
 static int
-test_cryptodev_caam_jr(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_caam_jr(void)
 {
 	gbl_driver_id =	rte_cryptodev_driver_id_get(
 			RTE_STR(CRYPTODEV_NAME_CAAM_JR_PMD));
@@ -14115,7 +14116,7 @@ test_cryptodev_bcmfs(void)
 }
 
 static int
-test_cryptodev_qat_raw_api(void /*argv __rte_unused, int argc __rte_unused*/)
+test_cryptodev_qat_raw_api(void)
 {
 	int ret;
 
diff --git a/app/test/test_cryptodev_aes_test_vectors.h b/app/test/test_cryptodev_aes_test_vectors.h
index c192d75..bb5f09f 100644
--- a/app/test/test_cryptodev_aes_test_vectors.h
+++ b/app/test/test_cryptodev_aes_test_vectors.h
@@ -2650,7 +2650,7 @@ static const struct blockcipher_test_case aes_cipheronly_test_cases[] = {
 		.op_mask = BLOCKCIPHER_TEST_OP_ENCRYPT,
 	},
 	{
-		.test_descr = "AES-192-CBC Encryption Scater gather",
+		.test_descr = "AES-192-CBC Encryption Scatter gather",
 		.test_data = &aes_test_data_10,
 		.op_mask = BLOCKCIPHER_TEST_OP_ENCRYPT,
 		.feature_mask = BLOCKCIPHER_TEST_FEATURE_SG |
diff --git a/app/test/test_eal_flags.c b/app/test/test_eal_flags.c
index 932fbe3..b4880ee 100644
--- a/app/test/test_eal_flags.c
+++ b/app/test/test_eal_flags.c
@@ -124,6 +124,7 @@ process_hugefiles(const char * prefix, enum hugepage_action action)
 			case HUGEPAGE_CHECK_EXISTS:
 				{
 					/* file exists, return */
+					closedir(hugepage_dir);
 					result = 1;
 					goto end;
 				}
diff --git a/app/test/test_event_crypto_adapter.c b/app/test/test_event_crypto_adapter.c
index 335211c..71b58fc 100644
--- a/app/test/test_event_crypto_adapter.c
+++ b/app/test/test_event_crypto_adapter.c
@@ -224,8 +224,7 @@ test_op_forward_mode(uint8_t session_less)
 		op->sess_type = RTE_CRYPTO_OP_SESSIONLESS;
 		first_xform = &cipher_xform;
 		sym_op->xform = first_xform;
-		uint32_t len = IV_OFFSET + MAXIMUM_IV_LENGTH +
-				(sizeof(struct rte_crypto_sym_xform) * 2);
+		uint32_t len = IV_OFFSET + MAXIMUM_IV_LENGTH;
 		op->private_data_offset = len;
 		/* Fill in private data information */
 		rte_memcpy(&m_data.response_info, &response_info,
@@ -419,8 +418,7 @@ test_op_new_mode(uint8_t session_less)
 		op->sess_type = RTE_CRYPTO_OP_SESSIONLESS;
 		first_xform = &cipher_xform;
 		sym_op->xform = first_xform;
-		uint32_t len = IV_OFFSET + MAXIMUM_IV_LENGTH +
-				(sizeof(struct rte_crypto_sym_xform) * 2);
+		uint32_t len = IV_OFFSET + MAXIMUM_IV_LENGTH;
 		op->private_data_offset = len;
 		/* Fill in private data information */
 		rte_memcpy(&m_data.response_info, &response_info,
@@ -516,7 +514,8 @@ configure_cryptodev(void)
 			NUM_MBUFS, MBUF_CACHE_SIZE,
 			DEFAULT_NUM_XFORMS *
 			sizeof(struct rte_crypto_sym_xform) +
-			MAXIMUM_IV_LENGTH,
+			MAXIMUM_IV_LENGTH +
+			sizeof(union rte_event_crypto_metadata),
 			rte_socket_id());
 	if (params.op_mpool == NULL) {
 		RTE_LOG(ERR, USER1, "Can't create CRYPTO_OP_POOL\n");
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index 47a7b19..5bb4786 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -2363,7 +2363,7 @@ test_pktmbuf_ext_shinfo_init_helper(struct rte_mempool *pktmbuf_pool)
 	if (rte_mbuf_refcnt_read(m) != 1)
 		GOTO_FAIL("%s: Invalid refcnt in mbuf\n", __func__);
 
-	buf_iova = rte_mempool_virt2iova(ext_buf_addr);
+	buf_iova = rte_mem_virt2iova(ext_buf_addr);
 	rte_pktmbuf_attach_extbuf(m, ext_buf_addr, buf_iova, buf_len,
 		ret_shinfo);
 	if (m->ol_flags != EXT_ATTACHED_MBUF)
diff --git a/app/test/test_power_cpufreq.c b/app/test/test_power_cpufreq.c
index 0c3adc5..94a3801 100644
--- a/app/test/test_power_cpufreq.c
+++ b/app/test/test_power_cpufreq.c
@@ -55,18 +55,20 @@ check_cur_freq(unsigned int lcore_id, uint32_t idx, bool turbo)
 	FILE *f;
 	char fullpath[PATH_MAX];
 	char buf[BUFSIZ];
+	enum power_management_env env;
 	uint32_t cur_freq;
+	uint32_t freq_conv;
 	int ret = -1;
 	int i;
 
 	if (snprintf(fullpath, sizeof(fullpath),
-		TEST_POWER_SYSFILE_SCALING_FREQ, lcore_id) < 0) {
+		TEST_POWER_SYSFILE_CPUINFO_FREQ, lcore_id) < 0) {
 		return 0;
 	}
 	f = fopen(fullpath, "r");
 	if (f == NULL) {
 		if (snprintf(fullpath, sizeof(fullpath),
-			TEST_POWER_SYSFILE_CPUINFO_FREQ, lcore_id) < 0) {
+			TEST_POWER_SYSFILE_SCALING_FREQ, lcore_id) < 0) {
 			return 0;
 		}
 		f = fopen(fullpath, "r");
@@ -80,15 +82,20 @@ check_cur_freq(unsigned int lcore_id, uint32_t idx, bool turbo)
 			goto fail_all;
 
 		cur_freq = strtoul(buf, NULL, TEST_POWER_CONVERT_TO_DECIMAL);
-
-		/* convert the frequency to nearest 100000 value
-		 * Ex: if cur_freq=1396789 then freq_conv=1400000
-		 * Ex: if cur_freq=800030 then freq_conv=800000
-		 */
-		unsigned int freq_conv = 0;
-		freq_conv = (cur_freq + TEST_FREQ_ROUNDING_DELTA)
-					/ TEST_ROUND_FREQ_TO_N_100000;
-		freq_conv = freq_conv * TEST_ROUND_FREQ_TO_N_100000;
+		freq_conv = cur_freq;
+
+		env = rte_power_get_env();
+
+		if (env == PM_ENV_PSTATE_CPUFREQ) {
+			/* convert the frequency to nearest 100000 value
+			 * Ex: if cur_freq=1396789 then freq_conv=1400000
+			 * Ex: if cur_freq=800030 then freq_conv=800000
+			 */
+			unsigned int freq_conv = 0;
+			freq_conv = (cur_freq + TEST_FREQ_ROUNDING_DELTA)
+						/ TEST_ROUND_FREQ_TO_N_100000;
+			freq_conv = freq_conv * TEST_ROUND_FREQ_TO_N_100000;
+		}
 
 		if (turbo)
 			ret = (freqs[idx] <= freq_conv ? 0 : -1);
diff --git a/buildtools/meson.build b/buildtools/meson.build
index 36161af..cfad51f 100644
--- a/buildtools/meson.build
+++ b/buildtools/meson.build
@@ -7,13 +7,13 @@ check_symbols = find_program('check-symbols.sh')
 ldflags_ibverbs_static = find_program('options-ibverbs-static.sh')
 binutils_avx512_check = find_program('binutils-avx512-check.sh')
 
-# set up map-to-win script using python, either built-in or external
 python3 = import('python').find_installation(required: false)
 if python3.found()
 	py3 = [python3]
 else
 	py3 = ['meson', 'runpython']
 endif
+echo = py3 + ['-c', 'import sys; print(*sys.argv[1:])']
 list_dir_globs = py3 + files('list-dir-globs.py')
 map_to_win_cmd = py3 + files('map_to_win.py')
 sphinx_wrapper = py3 + files('call-sphinx-build.py')
diff --git a/buildtools/symlink-drivers-solibs.py b/buildtools/symlink-drivers-solibs.py
new file mode 100644
index 0000000..9c99950
--- /dev/null
+++ b/buildtools/symlink-drivers-solibs.py
@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2021 Intel Corporation
+
+import os
+import sys
+import glob
+import shutil
+
+# post-install script for meson/ninja builds to symlink the PMDs stored in
+# $libdir/dpdk/pmds-*/ to $libdir. This is needed as some PMDs depend on
+# others, e.g. PCI device PMDs depending on the PCI bus driver.
+
+# parameters to script are paths relative to install prefix:
+# 1. directory for installed regular libs e.g. lib64
+# 2. subdirectory of libdir where the PMDs are
+# 3. directory for installed regular binaries e.g. bin
+
+os.chdir(os.environ['MESON_INSTALL_DESTDIR_PREFIX'])
+
+lib_dir = sys.argv[1]
+pmd_subdir = sys.argv[2]
+bin_dir = sys.argv[3]
+pmd_dir = os.path.join(lib_dir, pmd_subdir)
+
+# copy Windows PMDs to avoid any issues with symlinks since the
+# build could be a cross-compilation under WSL, Msys or Cygnus.
+# the filenames are dependent upon the specific toolchain in use.
+
+def copy_pmd_files(pattern, to_dir):
+	for file in glob.glob(os.path.join(pmd_dir, pattern)):
+		to = os.path.join(to_dir, os.path.basename(file))
+		shutil.copy2(file, to)
+		print(to + ' -> ' + file)
+
+copy_pmd_files('*rte_*.dll', bin_dir)
+copy_pmd_files('*rte_*.pdb', bin_dir)
+copy_pmd_files('*rte_*.lib', lib_dir)
+copy_pmd_files('*rte_*.dll.a', lib_dir)
+
+# symlink shared objects
+
+os.chdir(lib_dir)
+for file in glob.glob(os.path.join(pmd_subdir, 'librte_*.so*')):
+	to = os.path.basename(file)
+	if os.path.exists(to):
+		os.remove(to)
+	os.symlink(file, to)
+	print(to + ' -> ' + file)
diff --git a/config/meson.build b/config/meson.build
index 5b7439a..b2734fc 100644
--- a/config/meson.build
+++ b/config/meson.build
@@ -61,6 +61,10 @@ if not is_windows
 	meson.add_install_script('../buildtools/symlink-drivers-solibs.sh',
 			get_option('libdir'),
 			pmd_subdir_opt)
+elif meson.version().version_compare('>=0.55.0')
+	# 0.55.0 is required to use external program with add_install_script
+	meson.add_install_script(py3, '../buildtools/symlink-drivers-solibs.py',
+			get_option('libdir'), pmd_subdir_opt, get_option('bindir'))
 endif
 
 # set the machine type and cflags for it
diff --git a/debian/changelog b/debian/changelog
index ac0c210..490571c 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,14 @@
+dpdk (20.11.3-0ubuntu1) impish; urgency=medium
+
+  * Merge LTS stable release 20.11.3 (LP: #1940913)
+    Release notes are available at:
+    https://doc.dpdk.org/guides-20.11/rel_notes/release_20_11.html#id1
+    - Remove test-catch-coredumps.patch [now part of upstream]
+  * d/p/u/lp-1940957-net-i40e-support-25G-AOC-ACC-cables.patch: fix issues
+    with 25G AOC cables (LP: #1940957)
+
+ -- Christian Ehrhardt <[e-mail]>  Tue, 24 Aug 2021 12:28:59 +0200
+
 dpdk (20.11.2-1ubuntu3) impish; urgency=medium
 
   * No-change rebuild against libipsec-mb1
diff --git a/debian/patches/series b/debian/patches/series
index 37037a5..ef7c9a1 100644
--- a/debian/patches/series
+++ b/debian/patches/series
@@ -3,3 +3,4 @@ disable_lcores_autotest_ppc.patch
 disable_autopkgtest_fails.patch
 disable_armhf_autopkgtest_fails.patch
 disable_ppc64_autopkgtest_fails.patch
+ubuntu/lp-1940957-net-i40e-support-25G-AOC-ACC-cables.patch
diff --git a/debian/patches/ubuntu/lp-1940957-net-i40e-support-25G-AOC-ACC-cables.patch b/debian/patches/ubuntu/lp-1940957-net-i40e-support-25G-AOC-ACC-cables.patch
new file mode 100644
index 0000000..8606792
--- /dev/null
+++ b/debian/patches/ubuntu/lp-1940957-net-i40e-support-25G-AOC-ACC-cables.patch
@@ -0,0 +1,42 @@
+From f81d60e607bb8c13fcb13450e85d74721f1d9c8f Mon Sep 17 00:00:00 2001
+From: Yury Kylulin <[e-mail]>
+Date: Tue, 13 Apr 2021 18:29:50 +0300
+Subject: [PATCH] net/i40e: support 25G AOC/ACC cables
+
+[ upstream commit b1daa3461429e7674206a714c17adca65e9b44b4 ]
+
+Enable additional PHY types (25G-AOC and 25G-ACC) for set PHY config
+command.
+
+Ubuntu bug: https://bugs.launchpad.net/ubuntu/+source/dpdk/+bug/1940957
+Cc: [e-mail]
+
+Signed-off-by: Yury Kylulin <[e-mail]>
+Tested-by: Ashish Paul <[e-mail]>
+Signed-off-by: Christian Ehrhardt <[e-mail]>
+
+Origin: upstream, https://github.com/cpaelzer/dpdk-stable-queue/commit/f81d60e607bb8c13fcb13450e85d74721f1d9c8f
+Bug-Ubuntu: https://bugs.launchpad.net/bugs/1940957
+Last-Update: 2021-09-06
+
+---
+ drivers/net/i40e/i40e_ethdev.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
+index 508f90595f..0c896ea915 100644
+--- a/drivers/net/i40e/i40e_ethdev.c
++++ b/drivers/net/i40e/i40e_ethdev.c
+@@ -2264,7 +2264,8 @@ i40e_phy_conf_link(struct i40e_hw *hw,
+ 	phy_conf.phy_type = is_up ? cpu_to_le32(phy_type_mask) : 0;
+ 	phy_conf.phy_type_ext = is_up ? (I40E_AQ_PHY_TYPE_EXT_25G_KR |
+ 		I40E_AQ_PHY_TYPE_EXT_25G_CR | I40E_AQ_PHY_TYPE_EXT_25G_SR |
+-		I40E_AQ_PHY_TYPE_EXT_25G_LR) : 0;
++		I40E_AQ_PHY_TYPE_EXT_25G_LR | I40E_AQ_PHY_TYPE_EXT_25G_AOC |
++		I40E_AQ_PHY_TYPE_EXT_25G_ACC) : 0;
+ 	phy_conf.fec_config = phy_ab.fec_cfg_curr_mod_ext_info;
+ 	phy_conf.eee_capability = phy_ab.eee_capability;
+ 	phy_conf.eeer = phy_ab.eeer_val;
+-- 
+2.32.0
+
diff --git a/devtools/check-maintainers.sh b/devtools/check-maintainers.sh
index df3f740..71697bb 100755
--- a/devtools/check-maintainers.sh
+++ b/devtools/check-maintainers.sh
@@ -15,10 +15,10 @@ files () # <path> [<path> ...]
 	if [ -z "$1" ] ; then
 		return
 	fi
-	if [ -d .git ] ; then
+	if [ -r .git ] ; then
 		git ls-files "$1"
 	else
-		find "$1" -type f |
+		find $1 -type f |
 		sed 's,^\./,,'
 	fi |
 	# if not ended by /
diff --git a/doc/guides/contributing/coding_style.rst b/doc/guides/contributing/coding_style.rst
index bb3f3ef..a2a8130 100644
--- a/doc/guides/contributing/coding_style.rst
+++ b/doc/guides/contributing/coding_style.rst
@@ -55,7 +55,7 @@ License Header
 ~~~~~~~~~~~~~~
 
 Each file must begin with a special comment containing the
-`Software Package Data Exchange (SPDX) License Identfier <https://spdx.org/using-spdx-license-identifier>`_.
+`Software Package Data Exchange (SPDX) License Identifier <https://spdx.org/using-spdx-license-identifier>`_.
 
 Generally this is the BSD License, except for code granted special exceptions.
 The SPDX licences identifier is sufficient, a file should not contain
diff --git a/doc/guides/cryptodevs/scheduler.rst b/doc/guides/cryptodevs/scheduler.rst
index 835d999..d08207f 100644
--- a/doc/guides/cryptodevs/scheduler.rst
+++ b/doc/guides/cryptodevs/scheduler.rst
@@ -118,7 +118,7 @@ operation:
    than the designated threshold, otherwise it will be handled by the secondary
    worker.
 
-   A typical usecase in this mode is with the QAT cryptodev as the primary and
+   A typical use case in this mode is with the QAT cryptodev as the primary and
    a software cryptodev as the secondary worker. This may help applications to
    process additional crypto workload than what the QAT cryptodev can handle on
    its own, by making use of the available CPU cycles to deal with smaller
diff --git a/doc/guides/howto/pvp_reference_benchmark.rst b/doc/guides/howto/pvp_reference_benchmark.rst
index 553458d..484de3b 100644
--- a/doc/guides/howto/pvp_reference_benchmark.rst
+++ b/doc/guides/howto/pvp_reference_benchmark.rst
@@ -26,7 +26,7 @@ Setup overview
 
    PVP setup using 2 NICs
 
-In this diagram, each red arrow represents one logical core. This use-case
+In this diagram, each red arrow represents one logical core. This use case
 requires 6 dedicated logical cores. A forwarding configuration with a single
 NIC is also possible, requiring 3 logical cores.
 
diff --git a/doc/guides/nics/bnx2x.rst b/doc/guides/nics/bnx2x.rst
index 9ad4f9f..788a6da 100644
--- a/doc/guides/nics/bnx2x.rst
+++ b/doc/guides/nics/bnx2x.rst
@@ -105,7 +105,7 @@ Jumbo: Limitation
 -----------------
 
 Rx descriptor limit for number of segments per MTU is set to 1.
-PMD doesn't support Jumbo Rx scatter gather. Some applciations can
+PMD doesn't support Jumbo Rx scatter gather. Some applications can
 adjust mbuf_size based on this param and max_pkt_len.
 
 For others, PMD detects the condition where Rx packet length cannot
diff --git a/doc/guides/nics/dpaa.rst b/doc/guides/nics/dpaa.rst
index 917482d..7355ec3 100644
--- a/doc/guides/nics/dpaa.rst
+++ b/doc/guides/nics/dpaa.rst
@@ -297,7 +297,7 @@ FMC - FMAN Configuration Tool
 
 
    The details can be found in FMC Doc at:
-   `Frame Mnager Configuration Tool <https://www.nxp.com/docs/en/application-note/AN4760.pdf>`_.
+   `Frame Manager Configuration Tool <https://www.nxp.com/docs/en/application-note/AN4760.pdf>`_.
 
 FMLIB
 ~~~~~
@@ -307,7 +307,7 @@ FMLIB
 
    This is an alternate to the FMC based configuration. This library provides
    direct ioctl based interfaces for FMAN configuration as used by the FMC tool
-   as well. This helps in overcoming the main limitaiton of FMC - i.e. lack
+   as well. This helps in overcoming the main limitation of FMC - i.e. lack
    of dynamic configuration.
 
    The location for the fmd driver as used by FMLIB and FMC is as follows:
@@ -319,7 +319,7 @@ VSP (Virtual Storage Profile)
    The storage profiled are means to provide virtualized interface. A ranges of
    storage profiles cab be associated to Ethernet ports.
    They are selected during classification. Specify how the frame should be
-   written to memory and which buffer pool to select for packet storange in
+   written to memory and which buffer pool to select for packet storage in
    queues. Start and End margin of buffer can also be configured.
 
 Limitations
diff --git a/doc/guides/nics/ena.rst b/doc/guides/nics/ena.rst
index 0f1f63f..df72020 100644
--- a/doc/guides/nics/ena.rst
+++ b/doc/guides/nics/ena.rst
@@ -234,7 +234,7 @@ Example output:
 
    [...]
    EAL: PCI device 0000:00:06.0 on NUMA socket -1
-   EAL:   Invalid NUMA socket, default to 0
+   EAL: Device 0000:00:06.0 is not NUMA-aware, defaulting socket to 0
    EAL:   probe driver: 1d0f:ec20 net_ena
 
    Interactive-mode selected
diff --git a/doc/guides/nics/mlx5.rst b/doc/guides/nics/mlx5.rst
index 24d5a69..84e58fc 100644
--- a/doc/guides/nics/mlx5.rst
+++ b/doc/guides/nics/mlx5.rst
@@ -591,6 +591,13 @@ Driver options
   it is not recommended and may prevent NIC from sending packets over
   some configurations.
 
+  For ConnectX-4 and ConnectX-4 Lx NICs, automatically configured value
+  is insufficient for some traffic, because they require at least all L2 headers
+  to be inlined. For example, Q-in-Q adds 4 bytes to default 18 bytes
+  of Ethernet and VLAN, thus ``txq_inline_min`` must be set to 22.
+  MPLS would add 4 bytes per label. Final value must account for all possible
+  L2 encapsulation headers used in particular environment.
+
   Please, note, this minimal data inlining disengages eMPW feature (Enhanced
   Multi-Packet Write), because last one does not support partial packet inlining.
   This is not very critical due to minimal data inlining is mostly required
@@ -1281,7 +1288,7 @@ the DPDK application.
 
         echo -n "<device pci address" > /sys/bus/pci/drivers/mlx5_core/unbind
 
-5. Enbale switchdev mode::
+5. Enable switchdev mode::
 
         echo switchdev > /sys/class/net/<net device>/compat/devlink/mode
 
diff --git a/doc/guides/nics/octeontx2.rst b/doc/guides/nics/octeontx2.rst
index a4f2244..7b86cf5 100644
--- a/doc/guides/nics/octeontx2.rst
+++ b/doc/guides/nics/octeontx2.rst
@@ -153,7 +153,7 @@ Runtime Config Options
 
       -a 0002:02:00.0,max_sqb_count=64
 
-   With the above configuration, each send queue's decscriptor buffer count is
+   With the above configuration, each send queue's descriptor buffer count is
    limited to a maximum of 64 buffers.
 
 - ``Switch header enable`` (default ``none``)
@@ -242,7 +242,7 @@ configure the following features:
 #. Hierarchical scheduling
 #. Single rate - Two color, Two rate - Three color shaping
 
-Both DWRR and Static Priority(SP) hierarchial scheduling is supported.
+Both DWRR and Static Priority(SP) hierarchical scheduling is supported.
 
 Every parent can have atmost 10 SP Children and unlimited DWRR children.
 
diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
index aabd0f1..1f0aba3 100644
--- a/doc/guides/nics/virtio.rst
+++ b/doc/guides/nics/virtio.rst
@@ -509,7 +509,7 @@ are shown in below table:
    Split virtqueue in-order non-mergeable path  virtio_recv_pkts_inorder          virtio_xmit_pkts_inorder
    Split virtqueue vectorized Rx path           virtio_recv_pkts_vec              virtio_xmit_pkts
    Packed virtqueue mergeable path              virtio_recv_mergeable_pkts_packed virtio_xmit_pkts_packed
-   Packed virtqueue non-meregable path          virtio_recv_pkts_packed           virtio_xmit_pkts_packed
+   Packed virtqueue non-mergeable path          virtio_recv_pkts_packed           virtio_xmit_pkts_packed
    Packed virtqueue in-order mergeable path     virtio_recv_mergeable_pkts_packed virtio_xmit_pkts_packed
    Packed virtqueue in-order non-mergeable path virtio_recv_pkts_packed           virtio_xmit_pkts_packed
    Packed virtqueue vectorized Rx path          virtio_recv_pkts_packed_vec       virtio_xmit_pkts_packed
diff --git a/doc/guides/platform/dpaa.rst b/doc/guides/platform/dpaa.rst
index 20a0e39..3896929 100644
--- a/doc/guides/platform/dpaa.rst
+++ b/doc/guides/platform/dpaa.rst
@@ -78,7 +78,7 @@ compatible board:
    based config (if /tmp/fmc.bin is present). DPAA FMD will be used only if no
    previous fmc config is existing.
 
-   Note that fmlib based integratin rely on underlying fmd driver in kernel,
+   Note that fmlib based integration rely on underlying fmd driver in kernel,
    which is available as part of NXP kernel or NXP SDK.
 
 The following dependencies are not part of DPDK and must be installed
diff --git a/doc/guides/prog_guide/bbdev.rst b/doc/guides/prog_guide/bbdev.rst
index 6b2bd54..9619280 100644
--- a/doc/guides/prog_guide/bbdev.rst
+++ b/doc/guides/prog_guide/bbdev.rst
@@ -639,7 +639,7 @@ optionally the ``soft_output`` mbuf data pointers.
    "soft output","soft LLR output buffer (optional)"
    "op_flags","bitmask of all active operation capabilities"
    "rv_index","redundancy version index [0..3]"
-   "iter_max","maximum number of iterations to perofrm in decode all CBs"
+   "iter_max","maximum number of iterations to perform in decode all CBs"
    "iter_min","minimum number of iterations to perform in decoding all CBs"
    "iter_count","number of iterations to performed in decoding all CBs"
    "ext_scale","scale factor on extrinsic info (5 bits)"
diff --git a/doc/guides/prog_guide/env_abstraction_layer.rst b/doc/guides/prog_guide/env_abstraction_layer.rst
index 1f30e13..d124eb0 100644
--- a/doc/guides/prog_guide/env_abstraction_layer.rst
+++ b/doc/guides/prog_guide/env_abstraction_layer.rst
@@ -465,7 +465,7 @@ devices would fail anyway.
     - By default, the mempool, first asks for IOVA-contiguous memory using
       ``RTE_MEMZONE_IOVA_CONTIG``. This is slow in RTE_IOVA_PA mode and it may
       affect the application boot time.
-    - It is easy to enable large amount of IOVA-contiguous memory use-cases
+    - It is easy to enable large amount of IOVA-contiguous memory use cases
       with IOVA in VA mode.
 
     It is expected that all PCI drivers work in both RTE_IOVA_PA and
diff --git a/doc/guides/prog_guide/eventdev.rst b/doc/guides/prog_guide/eventdev.rst
index ccde086..347203f 100644
--- a/doc/guides/prog_guide/eventdev.rst
+++ b/doc/guides/prog_guide/eventdev.rst
@@ -120,7 +120,7 @@ Ports
 ~~~~~
 
 Ports are the points of contact between worker cores and the eventdev. The
-general use-case will see one CPU core using one port to enqueue and dequeue
+general use case will see one CPU core using one port to enqueue and dequeue
 events from an eventdev. Ports are linked to queues in order to retrieve events
 from those queues (more details in `Linking Queues and Ports`_ below).
 
diff --git a/doc/guides/prog_guide/multi_proc_support.rst b/doc/guides/prog_guide/multi_proc_support.rst
index 6b0ac30..815e8bd 100644
--- a/doc/guides/prog_guide/multi_proc_support.rst
+++ b/doc/guides/prog_guide/multi_proc_support.rst
@@ -325,7 +325,7 @@ supported. However, since sending messages (not requests) does not involve an
 IPC thread, sending messages while processing another message or request is
 supported.
 
-Since the memory sybsystem uses IPC internally, memory allocations and IPC must
+Since the memory subsystem uses IPC internally, memory allocations and IPC must
 not be mixed: it is not safe to use IPC inside a memory-related callback, nor is
 it safe to allocate/free memory inside IPC callbacks. Attempting to do so may
 lead to a deadlock.
diff --git a/doc/guides/prog_guide/qos_framework.rst b/doc/guides/prog_guide/qos_framework.rst
index 4e4ea33..e403b90 100644
--- a/doc/guides/prog_guide/qos_framework.rst
+++ b/doc/guides/prog_guide/qos_framework.rst
@@ -737,7 +737,7 @@ Strict priority scheduling of traffic classes within the same pipe is implemente
 which selects the queues in ascending order.
 Therefore, queue 0 (associated with TC 0, highest priority TC) is handled before
 queue 1 (TC 1, lower priority than TC 0),
-which is handled before queue 2 (TC 2, lower priority than TC 1) and it conitnues until queues of all TCs except the
+which is handled before queue 2 (TC 2, lower priority than TC 1) and it continues until queues of all TCs except the
 lowest priority TC are handled. At last, queues 12..15 (best effort TC, lowest priority TC) are handled.
 
 Upper Limit Enforcement
diff --git a/doc/guides/prog_guide/regexdev.rst b/doc/guides/prog_guide/regexdev.rst
index 3d8b591..5ca7e0c 100644
--- a/doc/guides/prog_guide/regexdev.rst
+++ b/doc/guides/prog_guide/regexdev.rst
@@ -124,7 +124,7 @@ The configuration mode is depended on the PMD capabilities.
 
 Online rule configuration is done using the following API functions:
 ``rte_regexdev_rule_db_update`` which add / remove rules from the rules
-precomplied list, and ``rte_regexdev_rule_db_compile_activate``
+precompiled list, and ``rte_regexdev_rule_db_compile_activate``
 which compile the rules and loads them to the RegEx HW.
 
 Offline rule configuration can be done by adding a pointer to the compiled
diff --git a/doc/guides/prog_guide/writing_efficient_code.rst b/doc/guides/prog_guide/writing_efficient_code.rst
index 7baeaae..a61e832 100644
--- a/doc/guides/prog_guide/writing_efficient_code.rst
+++ b/doc/guides/prog_guide/writing_efficient_code.rst
@@ -143,20 +143,21 @@ In order to achieve higher throughput,
 the DPDK attempts to aggregate the cost of processing each packet individually by processing packets in bursts.
 
 Using the testpmd application as an example,
-the burst size can be set on the command line to a value of 16 (also the default value).
-This allows the application to request 16 packets at a time from the PMD.
+the burst size can be set on the command line to a value of 32 (also the default value).
+This allows the application to request 32 packets at a time from the PMD.
 The testpmd application then immediately attempts to transmit all the packets that were received,
-in this case, all 16 packets.
+in this case, all 32 packets.
 
 The packets are not transmitted until the tail pointer is updated on the corresponding TX queue of the network port.
 This behavior is desirable when tuning for high throughput because
-the cost of tail pointer updates to both the RX and TX queues can be spread across 16 packets,
+the cost of tail pointer updates to both the RX and TX queues can be spread
+across 32 packets,
 effectively hiding the relatively slow MMIO cost of writing to the PCIe* device.
 However, this is not very desirable when tuning for low latency because
-the first packet that was received must also wait for another 15 packets to be received.
-It cannot be transmitted until the other 15 packets have also been processed because
+the first packet that was received must also wait for another 31 packets to be received.
+It cannot be transmitted until the other 31 packets have also been processed because
 the NIC will not know to transmit the packets until the TX tail pointer has been updated,
-which is not done until all 16 packets have been processed for transmission.
+which is not done until all 32 packets have been processed for transmission.
 
 To consistently achieve low latency, even under heavy system load,
 the application developer should avoid processing packets in bunches.
diff --git a/doc/guides/rawdevs/ioat.rst b/doc/guides/rawdevs/ioat.rst
index 250cfc4..59ba207 100644
--- a/doc/guides/rawdevs/ioat.rst
+++ b/doc/guides/rawdevs/ioat.rst
@@ -65,7 +65,7 @@ To assign an engine to a group::
         $ accel-config config-engine dsa0/engine0.1 --group-id=1
 
 To assign work queues to groups for passing descriptors to the engines a similar accel-config command can be used.
-However, the work queues also need to be configured depending on the use-case.
+However, the work queues also need to be configured depending on the use case.
 Some configuration options include:
 
 * mode (Dedicated/Shared): Indicates whether a WQ may accept jobs from multiple queues simultaneously.
diff --git a/doc/guides/rawdevs/ntb.rst b/doc/guides/rawdevs/ntb.rst
index 2c5fa76..2bb115d 100644
--- a/doc/guides/rawdevs/ntb.rst
+++ b/doc/guides/rawdevs/ntb.rst
@@ -17,7 +17,7 @@ some information by using scratchpad registers.
 BIOS setting on Intel Xeon
 --------------------------
 
-Intel Non-transparent Bridge needs special BIOS setting. The referencce for
+Intel Non-transparent Bridge needs special BIOS setting. The reference for
 Skylake is https://www.intel.com/content/dam/support/us/en/documents/server-products/Intel_Xeon_Processor_Scalable_Family_BIOS_User_Guide.pdf
 
 - Set the needed PCIe port as NTB to NTB mode on both hosts.
diff --git a/doc/guides/regexdevs/features_overview.rst b/doc/guides/regexdevs/features_overview.rst
index f90b394..c512bde 100644
--- a/doc/guides/regexdevs/features_overview.rst
+++ b/doc/guides/regexdevs/features_overview.rst
@@ -16,7 +16,7 @@ PCRE atomic grouping
   Support PCRE atomic grouping.
 
 PCRE back reference
-  Support PCRE back regerence.
+  Support PCRE back reference.
 
 PCRE back tracking ctrl
   Support PCRE back tracking ctrl.
diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst
index 2f498a0..c0d7efa 100644
--- a/doc/guides/rel_notes/deprecation.rst
+++ b/doc/guides/rel_notes/deprecation.rst
@@ -27,16 +27,18 @@ Deprecation Notices
 
 * rte_atomicNN_xxx: These APIs do not take memory order parameter. This does
   not allow for writing optimized code for all the CPU architectures supported
-  in DPDK. DPDK will adopt C11 atomic operations semantics and provide wrappers
-  using C11 atomic built-ins. These wrappers must be used for patches that
-  need to be merged in 20.08 onwards. This change will not introduce any
-  performance degradation.
+  in DPDK. DPDK has adopted the atomic operations from
+  https://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html. These
+  operations must be used for patches that need to be merged in 20.08 onwards.
+  This change will not introduce any performance degradation.
 
 * rte_smp_*mb: These APIs provide full barrier functionality. However, many
-  use cases do not require full barriers. To support such use cases, DPDK will
-  adopt C11 barrier semantics and provide wrappers using C11 atomic built-ins.
-  These wrappers must be used for patches that need to be merged in 20.08
-  onwards. This change will not introduce any performance degradation.
+  use cases do not require full barriers. To support such use cases, DPDK has
+  adopted atomic operations from
+  https://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html. These
+  operations and a new wrapper ``rte_atomic_thread_fence`` instead of
+  ``__atomic_thread_fence`` must be used for patches that need to be merged in
+  20.08 onwards. This change will not introduce any performance degradation.
 
 * lib: will fix extending some enum/define breaking the ABI. There are multiple
   samples in DPDK that enum/define terminated with a ``.*MAX.*`` value which is
@@ -127,12 +129,6 @@ Deprecation Notices
   from the release: ``0x16c8, 0x16c9, 0x16ca, 0x16ce, 0x16cf, 0x16df,``
   ``0x16d0, 0x16d1, 0x16d2, 0x16d4, 0x16d5, 0x16e7, 0x16e8, 0x16e9``.
 
-* sched: To allow more traffic classes, flexible mapping of pipe queues to
-  traffic classes, and subport level configuration of pipes and queues
-  changes will be made to macros, data structures and API functions defined
-  in "rte_sched.h". These changes are aligned to improvements suggested in the
-  RFC https://mails.dpdk.org/archives/dev/2018-November/120035.html.
-
 * metrics: The function ``rte_metrics_init`` will have a non-void return
   in order to notify errors instead of calling ``rte_exit``.
 
diff --git a/doc/guides/rel_notes/release_16_11.rst b/doc/guides/rel_notes/release_16_11.rst
index 92e0ec6..3cec914 100644
--- a/doc/guides/rel_notes/release_16_11.rst
+++ b/doc/guides/rel_notes/release_16_11.rst
@@ -77,7 +77,7 @@ New Features
   the current version, even 64 bytes packets take two slots with Virtio PMD on guest
   side.
 
-  The main impact is better performance for 0% packet loss use-cases, as it
+  The main impact is better performance for 0% packet loss use cases, as it
   behaves as if the virtqueue size was enlarged, so more packets can be buffered
   in the case of system perturbations. On the downside, small performance degradations
   were measured when running micro-benchmarks.
diff --git a/doc/guides/rel_notes/release_19_08.rst b/doc/guides/rel_notes/release_19_08.rst
index cbb27e8..d2baa82 100644
--- a/doc/guides/rel_notes/release_19_08.rst
+++ b/doc/guides/rel_notes/release_19_08.rst
@@ -151,7 +151,7 @@ New Features
   * Added multi-queue support to allow one af_xdp vdev with multiple netdev
     queues.
   * Enabled "need_wakeup" feature which can provide efficient support for the
-    usecase where the application and driver executing on the same core.
+    use case where the application and driver executing on the same core.
 
 * **Enabled infinite Rx in the PCAP PMD.**
 
diff --git a/doc/guides/rel_notes/release_20_11.rst b/doc/guides/rel_notes/release_20_11.rst
index 552c97e..b2f76bc 100644
--- a/doc/guides/rel_notes/release_20_11.rst
+++ b/doc/guides/rel_notes/release_20_11.rst
@@ -2109,3 +2109,343 @@ Tested Platforms
      Fixed in 21.08.
    * The hash value remains unchanged when the SCTP port value changed.
      Fixed in 21.08 new feature.
+
+20.11.3 Release Notes
+---------------------
+
+20.11.3 Fixes
+~~~~~~~~~~~~~
+
+* app/crypto-perf: fix out-of-place mempool allocation
+* app/test: fix IPv6 header initialization
+* app/testpmd: change port link speed without stopping all
+* app/testpmd: fix help string for port reset
+* app/testpmd: fix IPv4 checksum
+* app/testpmd: fix MAC address after port reset
+* app/testpmd: fix offloads for newly attached port
+* app/testpmd: fix Tx checksum calculation for tunnel
+* app/testpmd: fix type of FEC mode parsing output
+* bitmap: fix buffer overrun in bitmap init
+* build: support drivers symlink on Windows
+* bus: clarify log for non-NUMA-aware devices
+* bus/dpaa: fix freeing in FMAN interface destructor
+* bus/pci: fix IOVA as VA support for PowerNV
+* bus/pci: fix leak for unbound devices
+* common/mlx5: fix compatibility with OFED port query API
+* common/mlx5: fix memory region leak
+* common/mlx5: fix Netlink port name padding in probing
+* common/mlx5: fix Netlink receive message buffer size
+* common/mlx5: use new port query API if available
+* crypto/aesni_gcm: fix performance on some AVX512 CPUs
+* cryptodev: fix freeing after device release
+* crypto/mvsam: fix AES-GCM session parameters
+* crypto/mvsam: fix capabilities
+* crypto/mvsam: fix options parsing
+* crypto/mvsam: fix session data reset
+* crypto/octeontx2: fix IPsec session member overlap
+* crypto/octeontx2: fix lookaside IPsec IV pointer
+* crypto/octeontx: fix freeing after device release
+* crypto/qat: disable asymmetric crypto on GEN3
+* crypto/qat: fix Arm build with special memcpy
+* devtools: fix file listing in maintainers check
+* distributor: fix 128-bit write alignment
+* doc: add limitation for ConnectX-4 with L2 in mlx5 guide
+* doc: fix build on Windows with Meson 0.58
+* doc: fix default burst size in testpmd
+* doc: fix spelling
+* doc: fix typo in SPDX tag
+* doc: remove old deprecation notice for sched
+* doc: update atomic operation deprecation
+* drivers/net: fix memzone allocations for DMA memory
+* eal/windows: check callback parameter of alarm functions
+* eal/windows: cleanup virt2phys handle
+* ethdev: fix doc of flow action
+* eventdev: fix event port setup in Tx adapter
+* examples/l2fwd: fix [no-]mac-updating options
+* flow_classify: fix leaking rules on delete
+* graph: fix memory leak in stats
+* graph: fix null dereference in stats
+* ipc: stop mp control thread on cleanup
+* kni: fix crash on userspace VA for segmented packets
+* kni: fix mbuf allocation for kernel side use
+* malloc: fix size annotation for NUMA-aware realloc
+* mempool/octeontx2: fix shift calculation
+* net/bnxt: check access to possible null pointer
+* net/bnxt: cleanup code
+* net/bnxt: clear cached statistics
+* net/bnxt: detect bad opaque in Rx completion
+* net/bnxt: fix aarch32 build
+* net/bnxt: fix auto-negociation on Whitney+
+* net/bnxt: fix check for PTP support in FW
+* net/bnxt: fix error handling in VNIC prepare
+* net/bnxt: fix error messages in VNIC prepare
+* net/bnxt: fix missing barriers in completion handling
+* net/bnxt: fix nested lock during bonding
+* net/bnxt: fix null dereference in interrupt handler
+* net/bnxt: fix ring allocation and free
+* net/bnxt: fix ring and context memory allocation
+* net/bnxt: fix Rx burst size constraint
+* net/bnxt: fix Rx interrupt setting
+* net/bnxt: fix scalar Tx completion handling
+* net/bnxt: fix Tx descriptor status implementation
+* net/bnxt: fix typo in log message
+* net/bnxt: improve probing log message
+* net/bnxt: invoke device removal event on recovery failure
+* net/bnxt: remove unnecessary code
+* net/bnxt: remove unnecessary comment
+* net/bnxt: remove workaround for default VNIC
+* net/bnxt: set flow error after tunnel redirection free
+* net/bnxt: set flow error when free filter not available
+* net/bnxt: use common function to free VNIC resource
+* net/bnxt: workaround spurious zero stats in Thor
+* net/bonding: check flow setting
+* net/bonding: fix error message on flow verify
+* net/dpaa: fix headroom in VSP case
+* net/ena: enable multi-segment in Tx offload flags
+* net/ena: trigger reset on Tx prepare failure
+* net/hinic/base: fix LRO
+* net/hinic: fix MTU consistency with firmware
+* net/hinic: increase protection of the VLAN
+* net/hns3: fix Arm SVE build with GCC 8.3
+* net/hns3: fix delay for waiting to stop Rx/Tx
+* net/hns3: fix fake queue rollback
+* net/hns3: fix filter parsing comment
+* net/hns3: fix flow rule list in multi-process
+* net/hns3: fix maximum queues on configuration failure
+* net/hns3: fix residual MAC address entry
+* net/hns3: fix timing of clearing interrupt source
+* net/hns3: fix Tx prepare after stop
+* net/hns3: fix VLAN strip log
+* net/hns3: increase VF reset retry maximum
+* net/i40e: fix descriptor scan on Arm
+* net/i40e: fix flow director input set conflict
+* net/i40e: fix multi-process shared data
+* net/i40e: fix raw packet flow director
+* net/i40e: fix use after free in FDIR release
+* net/iavf: fix handling of unsupported promiscuous
+* net/iavf: fix RSS key access out of bound
+* net/iavf: fix scalar Rx
+* net/iavf: fix Tx threshold check
+* net/ice: fix data path in secondary process
+* net/ice: fix data path selection in secondary process
+* net/ice: fix default RSS key generation
+* net/ice: fix memzone leak when firmware is missing
+* net/ice: fix overflow in maximum packet length config
+* net/ixgbe: fix flow entry access after freeing
+* net/memif: fix abstract socket address length
+* net/mlx5: add Tx scheduling check on queue creation
+* net/mlx5: export PMD-specific API file
+* net/mlx5: fix default queue number in RSS flow rule
+* net/mlx5: fix flow engine type in function name
+* net/mlx5: fix imissed statistics
+* net/mlx5: fix indirect action modify rollback
+* net/mlx5: fix IPIP multi-tunnel validation
+* net/mlx5: fix match MPLS over GRE with key
+* net/mlx5: fix missing RSS expandable items
+* net/mlx5: fix missing RSS expansion of IPv6 frag
+* net/mlx5: fix MPLS RSS expansion
+* net/mlx5: fix multi-segment inline for the first segments
+* net/mlx5: fix overflow in mempool argument
+* net/mlx5: fix pattern expansion in RSS flow rules
+* net/mlx5: fix queue leaking in hairpin auto bind check
+* net/mlx5: fix representor interrupt handler
+* net/mlx5: fix RoCE LAG bond device probing
+* net/mlx5: fix RSS expansion for GTP
+* net/mlx5: fix RSS flow rule with L4 mismatch
+* net/mlx5: fix RSS pattern expansion
+* net/mlx5: fix r/w lock usage in DMA unmap
+* net/mlx5: fix Rx/Tx queue checks
+* net/mlx5: fix switchdev mode recognition
+* net/mlx5: fix threshold for mbuf replenishment in MPRQ
+* net/mlx5: fix timestamp initialization on empty clock queue
+* net/mlx5: fix TSO multi-segment inline length
+* net/mlx5: fix typo in vectorized Rx comments
+* net/mlx5: reject inner ethernet matching in GTP
+* net/mlx5: remove redundant operations in NEON Rx
+* net/mlx5: remove unsupported flow item MPLS over IP
+* net/mlx5: workaround drop action with old kernel
+* net/mvpp2: fix configured state dependency
+* net/mvpp2: fix port speed overflow
+* net/octeontx2: fix default MCAM allocation size
+* net/octeontx2: fix flow creation limit on CN98xx
+* net/octeontx2: fix TM node statistics query
+* net/octeontx2: use runtime LSO format indices
+* net/octeontx/base: fix debug build with clang
+* net/pfe: remove unnecessary null check
+* net/sfc: check ID overflow in action port ID
+* net/sfc: fix aarch32 build
+* net/sfc: fix MAC stats lock in xstats query by ID
+* net/sfc: fix MAC stats update for stopped device
+* net/sfc: fix outer L4 checksum Rx
+* net/sfc: fix outer match in MAE backend
+* net/sfc: fix reading adapter state without locking
+* net/sfc: fix xstats query by ID according to ethdev
+* net/sfc: fix xstats query by unsorted list of IDs
+* net/softnic: fix connection memory leak
+* net/softnic: fix memory leak as profile is freed
+* net/softnic: fix memory leak in arguments parsing
+* net/softnic: fix null dereference in arguments parsing
+* net/tap: fix Rx checksum flags on IP options packets
+* net/tap: fix Rx checksum flags on TCP packets
+* net/virtio: fix aarch32 build
+* net/virtio: fix default duplex mode
+* net/virtio: fix interrupt handle leak
+* net/virtio: fix refill order in packed ring datapath
+* net/virtio: fix Rx scatter offload
+* net/virtio: report maximum MTU in device info
+* raw/ioat: fix config script queue size calculation
+* regex/mlx5: fix redundancy in device removal
+* regex/mlx5: fix size of setup constants
+* rib: fix max depth IPv6 lookup
+* sched: fix profile allocation failure handling
+* sched: rework configuration failure handling
+* table: fix bucket empty check
+* test/crypto: fix autotest function parameters
+* test/crypto: fix mbuf reset after null check
+* test/crypto: fix mempool size for session-less
+* test/crypto: fix typo in AES case
+* test/crypto: fix typo in ESN case
+* test/mbuf: fix virtual address conversion
+* test/power: fix CPU frequency check for intel_pstate
+* test/power: fix CPU frequency when turbo enabled
+* tests/cmdline: fix memory leaks
+* tests/eal: fix memory leak
+* vdpa/mlx5: fix overflow in queue attribute
+* vdpa/mlx5: fix TSO offload without checksum
+* version: 20.11.3-rc1
+* vfio: add stdbool include
+* vhost: check header for legacy dequeue offload
+* vhost/crypto: check request pointer before dereference
+* vhost: fix crash on reconnect
+* vhost: fix lock on device readiness notification
+* vhost: fix missing guest pages table NUMA realloc
+* vhost: fix missing memory table NUMA realloc
+* vhost: fix NUMA reallocation with multi-queue
+
+20.11.3 Validation
+~~~~~~~~~~~~~~~~~~
+
+* Intel(R) Testing
+
+   * Basic Intel(R) NIC(ixgbe, i40e and ice) testing
+      * PF (i40e)
+      * PF (ixgbe)
+      * PF (ice)
+      * VF (i40e)
+      * VF (ixgbe)
+      * VF (ice)
+      * Compile Testing
+      * Intel NIC single core/NIC performance
+      * Power and IPsec
+
+   * Basic cryptodev and virtio testing
+
+      * vhost/virtio basic loopback, PVP and performance test
+      * cryptodev Function/Performance
+
+
+* Nvidia(R) Testing
+
+   * Basic functionality with testpmd
+
+      * Tx/Rx
+      * xstats
+      * Timestamps
+      * Link status
+      * RTE flow and flow_director
+      * RSS
+      * VLAN stripping and insertion
+      * Checksum/TSO
+      * ptype
+      * link_status_interrupt example application
+      * l3fwd-power example application
+      * Multi-process example applications
+      * Hardware LRO tests
+
+   * Build tests
+
+      * Ubuntu 20.04.2 with MLNX_OFED_LINUX-5.4-1.0.3.0.
+      * Ubuntu 20.04.2 with rdma-core master (64d1ae5).
+      * Ubuntu 20.04.2 with rdma-core v28.0.
+      * Ubuntu 18.04.5 with rdma-core v17.1.
+      * Ubuntu 18.04.5 with rdma-core master (5b0f5b2) (i386).
+      * Ubuntu 16.04.7 with rdma-core v22.7.
+      * Fedora 34 with rdma-core v36.0.
+      * Fedora 36 (Rawhide) with rdma-core v36.0 (only with gcc).
+      * CentOS 7 7.9.2009 with rdma-core master (64d1ae5).
+      * CentOS 7 7.9.2009 with MLNX_OFED_LINUX-5.4-1.0.3.0.
+      * CentOS 8 8.3.2011 with rdma-core master (64d1ae5).
+      * OpenSUSE Leap 15.3 with rdma-core v31.0.
+
+   * ConnectX-5
+
+      * Ubuntu 20.04
+      * Driver MLNX_OFED_LINUX-5.4-1.0.3.0
+      * Kernel: 5.14.0-rc6 / Driver: rdma-core v36.0
+      * fw 16.31.1014
+
+   * ConnectX-4 Lx
+
+      * Ubuntu 20.04
+      * Driver MLNX_OFED_LINUX-5.4-1.0.3.0
+      * Kernel: 5.14.0-rc6 / Driver: rdma-core v36.0
+      * fw 14.31.1014
+
+
+* Red Hat(R) Testing
+
+   * Platform
+
+      * RHEL 8
+      * Kernel 4.18
+      * Qemu 6.0
+      * X540-AT2 NIC(ixgbe, 10G)
+
+   * Functionality
+
+      * Guest with device assignment(PF) throughput testing(1G hugepage size)
+      * Guest with device assignment(PF) throughput testing(2M hugepage size)
+      * Guest with device assignment(VF) throughput testing
+      * PVP (host dpdk testpmd as vswitch) 1Q: throughput testing
+      * PVP vhost-user 2Q throughput testing
+      * PVP vhost-user 1Q cross numa node  throughput testing
+      * Guest with vhost-user 2 queues throughput testing
+      * vhost-user reconnect with dpdk-client, qemu-server: qemu reconnect
+      * vhost-user reconnect with dpdk-client, qemu-server: ovs reconnect
+      * PVP 1Q live migration testing
+      * PVP 1Q cross numa node live migration testing
+      * Guest with ovs+dpdk+vhost-user 1Q live migration testing
+      * Guest with ovs+dpdk+vhost-user 1Q live migration testing (2M)
+      * Guest with ovs+dpdk+vhost-user 2Q live migration testing
+      * Host PF + DPDK testing
+      * Host VF + DPDK testing
+
+* Canonical(R) Testing
+
+   * Build tests of DPDK & OVS 2.15.0 on Ubuntu 21.04 (meson based)
+   * Functional and performance tests based on OVS-DPDK on x86_64
+   * Autopkgtests for DPDK and OpenvSwitch
+
+20.11.3 Known Issues
+~~~~~~~~~~~~~~~~~~~~
+
+* ICE
+
+   * creating 512 acl rules after creating a full mask switch rule fails.
+
+* vhost/virtio
+
+   * udp-fragmentation-offload cannot be setup on Ubuntu 19.10 VMs.
+     https://bugzilla.kernel.org/show_bug.cgi?id=207075
+   * vm2vm virtio-net connectivity between two vms randomly fails due
+     to lost connection after vhost reconnect.
+
+* unit tests
+
+   * unit_tests_power/power_cpufreq fails.
+     https://bugs.dpdk.org/show_bug.cgi?id=790
+
+* IAVF
+
+   * cvl_advanced_iavf_rss: after changing the SCTP port value, the hash value
+     remains unchanged.
diff --git a/doc/guides/rel_notes/release_2_2.rst b/doc/guides/rel_notes/release_2_2.rst
index cea5c87..8273473 100644
--- a/doc/guides/rel_notes/release_2_2.rst
+++ b/doc/guides/rel_notes/release_2_2.rst
@@ -322,7 +322,7 @@ Drivers
 
   Several customers have reported a link flap issue on 82579. The symptoms
   are random and intermittent link losses when 82579 is connected to specific
-  switches. the Issue was root caused as an inter-operability problem between
+  switches. the Issue was root caused as an interoperability problem between
   the NIC and at least some Broadcom PHYs in the Energy Efficient Ethernet
   wake mechanism.
 
diff --git a/doc/guides/sample_app_ug/fips_validation.rst b/doc/guides/sample_app_ug/fips_validation.rst
index ca37fc0..56df434 100644
--- a/doc/guides/sample_app_ug/fips_validation.rst
+++ b/doc/guides/sample_app_ug/fips_validation.rst
@@ -113,7 +113,7 @@ where,
   * mbuf-dataroom: By default the application creates mbuf pool with maximum
     possible data room (65535 bytes). If the user wants to test scatter-gather
     list feature of the PMD he or she may set this value to reduce the dataroom
-    size so that the input data may be dividied into multiple chained mbufs.
+    size so that the input data may be divided into multiple chained mbufs.
 
 
 To run the application in linux environment to test one AES FIPS test data
diff --git a/doc/guides/sample_app_ug/hello_world.rst b/doc/guides/sample_app_ug/hello_world.rst
index 7cb9279..6ec93e0 100644
--- a/doc/guides/sample_app_ug/hello_world.rst
+++ b/doc/guides/sample_app_ug/hello_world.rst
@@ -1,4 +1,4 @@
-o..  SPDX-License-Identifier: BSD-3-Clause
+..  SPDX-License-Identifier: BSD-3-Clause
     Copyright(c) 2010-2014 Intel Corporation.
 
 Hello World Sample Application
diff --git a/doc/guides/sample_app_ug/ipsec_secgw.rst b/doc/guides/sample_app_ug/ipsec_secgw.rst
index 176e292..c07275e 100644
--- a/doc/guides/sample_app_ug/ipsec_secgw.rst
+++ b/doc/guides/sample_app_ug/ipsec_secgw.rst
@@ -93,7 +93,7 @@ Additionally the event mode introduces two submodes of processing packets:
   protocol use case, the worker thread resembles l2fwd worker thread as the IPsec
   processing is done entirely in HW. This mode can be used to benchmark the raw
   performance of the HW. The driver submode is selected with --single-sa option
-  (used also by poll mode). When --single-sa option is used in conjution with event
+  (used also by poll mode). When --single-sa option is used in conjunction with event
   mode then index passed to --single-sa is ignored.
 
 * App submode: This submode has all the features currently implemented with the
diff --git a/doc/guides/sample_app_ug/performance_thread.rst b/doc/guides/sample_app_ug/performance_thread.rst
index 4c6a1db..9b09838 100644
--- a/doc/guides/sample_app_ug/performance_thread.rst
+++ b/doc/guides/sample_app_ug/performance_thread.rst
@@ -1176,7 +1176,7 @@ Tracing of events can be individually masked, and the mask may be programmed
 at run time. An unmasked event results in a callback that provides information
 about the event. The default callback simply prints trace information. The
 default mask is 0 (all events off) the mask can be modified by calling the
-function ``lthread_diagniostic_set_mask()``.
+function ``lthread_diagnostic_set_mask()``.
 
 It is possible register a user callback function to implement more
 sophisticated diagnostic functions.
diff --git a/doc/guides/testpmd_app_ug/run_app.rst b/doc/guides/testpmd_app_ug/run_app.rst
index 6f9ff13..d76eb10 100644
--- a/doc/guides/testpmd_app_ug/run_app.rst
+++ b/doc/guides/testpmd_app_ug/run_app.rst
@@ -112,7 +112,7 @@ The command line options are:
     Set the data size of the mbufs used to N bytes, where N < 65536.
     The default value is 2048. If multiple mbuf-size values are specified the
     extra memory pools will be created for allocating mbufs to receive packets
-    with buffer splittling features.
+    with buffer splitting features.
 
 *   ``--total-num-mbufs=N``
 
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 3187756..a004d89 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -1732,7 +1732,7 @@ List all items from the ptype mapping table::
 
 Where:
 
-* ``valid_only``: A flag indicates if only list valid items(=1) or all itemss(=0).
+* ``valid_only``: A flag indicates if only list valid items(=1) or all items(=0).
 
 Replace a specific or a group of software defined ptype with a new one::
 
@@ -4767,7 +4767,7 @@ Sample Raw encapsulation rule
 
 Raw encapsulation configuration can be set by the following commands
 
-Eecapsulating VxLAN::
+Encapsulating VxLAN::
 
  testpmd> set raw_encap 4 eth src is 10:11:22:33:44:55 / vlan tci is 1
         inner_type is 0x0800 / ipv4 / udp dst is 4789 / vxlan vni
diff --git a/doc/guides/tools/hugepages.rst b/doc/guides/tools/hugepages.rst
index 6d3f410..dd24c80 100644
--- a/doc/guides/tools/hugepages.rst
+++ b/doc/guides/tools/hugepages.rst
@@ -62,7 +62,7 @@ Options
 
 .. warning::
 
-   While any user can run the ``dpdk-hugpages.py`` script to view the
+   While any user can run the ``dpdk-hugepages.py`` script to view the
    status of huge pages, modifying the setup requires root privileges.
 
 
@@ -71,8 +71,8 @@ Examples
 
 To display current huge page settings::
 
-   dpdk-hugpages.py -s
+   dpdk-hugepages.py -s
 
 To a complete setup of with 2 Gigabyte of 1G huge pages::
 
-   dpdk-hugpages.py -p 1G --setup 2G
+   dpdk-hugepages.py -p 1G --setup 2G
diff --git a/doc/meson.build b/doc/meson.build
index c5410d8..d6cf85a 100644
--- a/doc/meson.build
+++ b/doc/meson.build
@@ -11,5 +11,5 @@ if doc_targets.length() == 0
 else
 	message = 'Building docs:'
 endif
-run_target('doc', command: ['echo', message, doc_target_names],
+run_target('doc', command: [echo, message, doc_target_names],
 	depends: doc_targets)
diff --git a/drivers/bus/dpaa/base/fman/fman.c b/drivers/bus/dpaa/base/fman/fman.c
index 39102bc..997c94f 100644
--- a/drivers/bus/dpaa/base/fman/fman.c
+++ b/drivers/bus/dpaa/base/fman/fman.c
@@ -50,7 +50,7 @@ if_destructor(struct __fman_if *__if)
 		free(bp);
 	}
 cleanup:
-	free(__if);
+	rte_free(__if);
 }
 
 static int
diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
index 2e1808b..e8d1faa 100644
--- a/drivers/bus/pci/linux/pci.c
+++ b/drivers/bus/pci/linux/pci.c
@@ -331,7 +331,7 @@ pci_scan_one(const char *dirname, const struct rte_pci_addr *addr)
 		else
 			dev->kdrv = RTE_PCI_KDRV_UNKNOWN;
 	} else {
-		dev->kdrv = RTE_PCI_KDRV_NONE;
+		free(dev);
 		return 0;
 	}
 	/* device is valid, add in list (sorted) */
@@ -569,7 +569,7 @@ pci_device_iommu_support_va(__rte_unused const struct rte_pci_device *dev)
 
 	/* Check for a PowerNV platform */
 	while (getline(&line, &len, fp) != -1) {
-		if (strstr(line, "platform") != NULL)
+		if (strstr(line, "platform") == NULL)
 			continue;
 
 		if (strstr(line, "PowerNV") != NULL) {
diff --git a/drivers/bus/pci/pci_common.c b/drivers/bus/pci/pci_common.c
index 9b8d769..fa887de 100644
--- a/drivers/bus/pci/pci_common.c
+++ b/drivers/bus/pci/pci_common.c
@@ -16,6 +16,7 @@
 #include <rte_bus.h>
 #include <rte_pci.h>
 #include <rte_bus_pci.h>
+#include <rte_lcore.h>
 #include <rte_per_lcore.h>
 #include <rte_memory.h>
 #include <rte_eal.h>
@@ -190,7 +191,9 @@ rte_pci_probe_one_driver(struct rte_pci_driver *dr,
 	}
 
 	if (dev->device.numa_node < 0) {
-		RTE_LOG(WARNING, EAL, "  Invalid NUMA socket, default to 0\n");
+		if (rte_socket_count() > 1)
+			RTE_LOG(INFO, EAL, "Device %s is not NUMA-aware, defaulting socket to 0\n",
+					dev->name);
 		dev->device.numa_node = 0;
 	}
 
diff --git a/drivers/bus/vmbus/vmbus_common.c b/drivers/bus/vmbus/vmbus_common.c
index 39b3308..09b8c3c 100644
--- a/drivers/bus/vmbus/vmbus_common.c
+++ b/drivers/bus/vmbus/vmbus_common.c
@@ -15,6 +15,7 @@
 #include <rte_eal.h>
 #include <rte_tailq.h>
 #include <rte_devargs.h>
+#include <rte_lcore.h>
 #include <rte_malloc.h>
 #include <rte_errno.h>
 #include <rte_memory.h>
@@ -112,7 +113,9 @@ vmbus_probe_one_driver(struct rte_vmbus_driver *dr,
 	dev->driver = dr;
 
 	if (dev->device.numa_node < 0) {
-		VMBUS_LOG(WARNING, "  Invalid NUMA socket, default to 0");
+		if (rte_socket_count() > 1)
+			VMBUS_LOG(INFO, "Device %s is not NUMA-aware, defaulting socket to 0",
+					guid);
 		dev->device.numa_node = 0;
 	}
 
diff --git a/drivers/common/iavf/iavf_impl.c b/drivers/common/iavf/iavf_impl.c
index fc0da31..f80878b 100644
--- a/drivers/common/iavf/iavf_impl.c
+++ b/drivers/common/iavf/iavf_impl.c
@@ -6,7 +6,6 @@
 #include <inttypes.h>
 
 #include <rte_common.h>
-#include <rte_random.h>
 #include <rte_malloc.h>
 #include <rte_memzone.h>
 
@@ -19,13 +18,15 @@ iavf_allocate_dma_mem_d(__rte_unused struct iavf_hw *hw,
 			u64 size,
 			u32 alignment)
 {
+	static uint64_t iavf_dma_memzone_id;
 	const struct rte_memzone *mz = NULL;
 	char z_name[RTE_MEMZONE_NAMESIZE];
 
 	if (!mem)
 		return IAVF_ERR_PARAM;
 
-	snprintf(z_name, sizeof(z_name), "iavf_dma_%"PRIu64, rte_rand());
+	snprintf(z_name, sizeof(z_name), "iavf_dma_%" PRIu64,
+		__atomic_fetch_add(&iavf_dma_memzone_id, 1, __ATOMIC_RELAXED));
 	mz = rte_memzone_reserve_bounded(z_name, size, SOCKET_ID_ANY,
 					 RTE_MEMZONE_IOVA_CONTIG, alignment,
 					 RTE_PGSIZE_2M);
diff --git a/drivers/common/mlx5/linux/meson.build b/drivers/common/mlx5/linux/meson.build
index fa9686f..c2d580c 100644
--- a/drivers/common/mlx5/linux/meson.build
+++ b/drivers/common/mlx5/linux/meson.build
@@ -94,6 +94,10 @@ has_sym_args = [
 	'IBV_WQ_FLAG_RX_END_PADDING' ],
 	[ 'HAVE_MLX5DV_DR_DEVX_PORT', 'infiniband/mlx5dv.h',
 	'mlx5dv_query_devx_port' ],
+	[ 'HAVE_MLX5DV_DR_DEVX_PORT_V35', 'infiniband/mlx5dv.h',
+	'mlx5dv_query_port' ],
+	[ 'HAVE_MLX5DV_DR_CREATE_DEST_IB_PORT', 'infiniband/mlx5dv.h',
+	'mlx5dv_dr_action_create_dest_ib_port' ],
 	[ 'HAVE_IBV_DEVX_OBJ', 'infiniband/mlx5dv.h',
 	'mlx5dv_devx_obj_create' ],
 	[ 'HAVE_IBV_FLOW_DEVX_COUNTERS', 'infiniband/mlx5dv.h',
diff --git a/drivers/common/mlx5/linux/mlx5_glue.c b/drivers/common/mlx5/linux/mlx5_glue.c
index 964f7e7..09fdce1 100644
--- a/drivers/common/mlx5/linux/mlx5_glue.c
+++ b/drivers/common/mlx5/linux/mlx5_glue.c
@@ -391,7 +391,7 @@ mlx5_glue_dr_create_flow_action_dest_flow_tbl(void *tbl)
 static void *
 mlx5_glue_dr_create_flow_action_dest_port(void *domain, uint32_t port)
 {
-#ifdef HAVE_MLX5DV_DR_DEVX_PORT
+#ifdef HAVE_MLX5DV_DR_CREATE_DEST_IB_PORT
 	return mlx5dv_dr_action_create_dest_ib_port(domain, port);
 #else
 #ifdef HAVE_MLX5DV_DR_ESWITCH
@@ -1087,17 +1087,54 @@ mlx5_glue_devx_wq_query(struct ibv_wq *wq, const void *in, size_t inlen,
 static int
 mlx5_glue_devx_port_query(struct ibv_context *ctx,
 			  uint32_t port_num,
-			  struct mlx5dv_devx_port *mlx5_devx_port)
-{
+			  struct mlx5_port_info *info)
+{
+	int err = 0;
+
+	info->query_flags = 0;
+#ifdef HAVE_MLX5DV_DR_DEVX_PORT_V35
+	/* The DevX port query API is implemented (rdma-core v35 and above). */
+	struct mlx5_ib_uapi_query_port devx_port;
+
+	memset(&devx_port, 0, sizeof(devx_port));
+	err = mlx5dv_query_port(ctx, port_num, &devx_port);
+	if (err)
+		return err;
+	if (devx_port.flags & MLX5DV_QUERY_PORT_VPORT_REG_C0) {
+		info->vport_meta_tag = devx_port.reg_c0.value;
+		info->vport_meta_mask = devx_port.reg_c0.mask;
+		info->query_flags |= MLX5_PORT_QUERY_REG_C0;
+	}
+	if (devx_port.flags & MLX5DV_QUERY_PORT_VPORT) {
+		info->vport_id = devx_port.vport;
+		info->query_flags |= MLX5_PORT_QUERY_VPORT;
+	}
+#else
 #ifdef HAVE_MLX5DV_DR_DEVX_PORT
-	return mlx5dv_query_devx_port(ctx, port_num, mlx5_devx_port);
+	/* The legacy DevX port query API is implemented (prior v35). */
+	struct mlx5dv_devx_port devx_port = {
+		.comp_mask = MLX5DV_DEVX_PORT_VPORT |
+			     MLX5DV_DEVX_PORT_MATCH_REG_C_0
+	};
+
+	err = mlx5dv_query_devx_port(ctx, port_num, &devx_port);
+	if (err)
+		return err;
+	if (devx_port.comp_mask & MLX5DV_DEVX_PORT_MATCH_REG_C_0) {
+		info->vport_meta_tag = devx_port.reg_c_0.value;
+		info->vport_meta_mask = devx_port.reg_c_0.mask;
+		info->query_flags |= MLX5_PORT_QUERY_REG_C0;
+	}
+	if (devx_port.comp_mask & MLX5DV_DEVX_PORT_VPORT) {
+		info->vport_id = devx_port.vport_num;
+		info->query_flags |= MLX5_PORT_QUERY_VPORT;
+	}
 #else
-	(void)ctx;
-	(void)port_num;
-	(void)mlx5_devx_port;
-	errno = ENOTSUP;
-	return errno;
-#endif
+	RTE_SET_USED(ctx);
+	RTE_SET_USED(port_num);
+#endif /* HAVE_MLX5DV_DR_DEVX_PORT */
+#endif /* HAVE_MLX5DV_DR_DEVX_PORT_V35 */
+	return err;
 }
 
 static int
diff --git a/drivers/common/mlx5/linux/mlx5_glue.h b/drivers/common/mlx5/linux/mlx5_glue.h
index 9e385be..f08c837 100644
--- a/drivers/common/mlx5/linux/mlx5_glue.h
+++ b/drivers/common/mlx5/linux/mlx5_glue.h
@@ -84,6 +84,20 @@ struct mlx5dv_dr_action;
 struct mlx5dv_devx_port;
 #endif
 
+#ifndef HAVE_MLX5DV_DR_DEVX_PORT_V35
+struct mlx5dv_port;
+#endif
+
+#define MLX5_PORT_QUERY_VPORT (1u << 0)
+#define MLX5_PORT_QUERY_REG_C0 (1u << 1)
+
+struct mlx5_port_info {
+	uint16_t query_flags;
+	uint16_t vport_id; /* Associated VF vport index (if any). */
+	uint32_t vport_meta_tag; /* Used for vport index match ove VF LAG. */
+	uint32_t vport_meta_mask; /* Used for vport index field match mask. */
+};
+
 #ifndef HAVE_MLX5_DR_CREATE_ACTION_FLOW_METER
 struct mlx5dv_dr_flow_meter_attr;
 #endif
@@ -311,7 +325,7 @@ struct mlx5_glue {
 			     void *out, size_t outlen);
 	int (*devx_port_query)(struct ibv_context *ctx,
 			       uint32_t port_num,
-			       struct mlx5dv_devx_port *mlx5_devx_port);
+			       struct mlx5_port_info *info);
 	int (*dr_dump_domain)(FILE *file, void *domain);
 	int (*devx_query_eqn)(struct ibv_context *context, uint32_t cpus,
 			      uint32_t *eqn);
diff --git a/drivers/common/mlx5/linux/mlx5_nl.c b/drivers/common/mlx5/linux/mlx5_nl.c
index ef7a521..1f765dc 100644
--- a/drivers/common/mlx5/linux/mlx5_nl.c
+++ b/drivers/common/mlx5/linux/mlx5_nl.c
@@ -33,6 +33,8 @@
 #define MLX5_SEND_BUF_SIZE 32768
 /* Receive buffer size for the Netlink socket */
 #define MLX5_RECV_BUF_SIZE 32768
+/* Maximal physical port name length. */
+#define MLX5_PHYS_PORT_NAME_MAX 128
 
 /** Parameters of VLAN devices created by driver. */
 #define MLX5_VMWA_VLAN_DEVICE_PFX "evmlx"
@@ -187,8 +189,8 @@ int
 mlx5_nl_init(int protocol)
 {
 	int fd;
-	int sndbuf_size = MLX5_SEND_BUF_SIZE;
-	int rcvbuf_size = MLX5_RECV_BUF_SIZE;
+	int buf_size;
+	socklen_t opt_size;
 	struct sockaddr_nl local = {
 		.nl_family = AF_NETLINK,
 	};
@@ -199,16 +201,36 @@ mlx5_nl_init(int protocol)
 		rte_errno = errno;
 		return -rte_errno;
 	}
-	ret = setsockopt(fd, SOL_SOCKET, SO_SNDBUF, &sndbuf_size, sizeof(int));
+	opt_size = sizeof(buf_size);
+	ret = getsockopt(fd, SOL_SOCKET, SO_SNDBUF, &buf_size, &opt_size);
 	if (ret == -1) {
 		rte_errno = errno;
 		goto error;
 	}
-	ret = setsockopt(fd, SOL_SOCKET, SO_RCVBUF, &rcvbuf_size, sizeof(int));
+	DRV_LOG(DEBUG, "Netlink socket send buffer: %d", buf_size);
+	if (buf_size < MLX5_SEND_BUF_SIZE) {
+		ret = setsockopt(fd, SOL_SOCKET, SO_SNDBUF,
+				 &buf_size, sizeof(buf_size));
+		if (ret == -1) {
+			rte_errno = errno;
+			goto error;
+		}
+	}
+	opt_size = sizeof(buf_size);
+	ret = getsockopt(fd, SOL_SOCKET, SO_RCVBUF, &buf_size, &opt_size);
 	if (ret == -1) {
 		rte_errno = errno;
 		goto error;
 	}
+	DRV_LOG(DEBUG, "Netlink socket recv buffer: %d", buf_size);
+	if (buf_size < MLX5_RECV_BUF_SIZE) {
+		ret = setsockopt(fd, SOL_SOCKET, SO_RCVBUF,
+				 &buf_size, sizeof(buf_size));
+		if (ret == -1) {
+			rte_errno = errno;
+			goto error;
+		}
+	}
 	ret = bind(fd, (struct sockaddr *)&local, sizeof(local));
 	if (ret == -1) {
 		rte_errno = errno;
@@ -330,11 +352,7 @@ mlx5_nl_recv(int nlsk_fd, uint32_t sn, int (*cb)(struct nlmsghdr *, void *arg),
 	     void *arg)
 {
 	struct sockaddr_nl sa;
-	void *buf = mlx5_malloc(0, MLX5_RECV_BUF_SIZE, 0, SOCKET_ID_ANY);
-	struct iovec iov = {
-		.iov_base = buf,
-		.iov_len = MLX5_RECV_BUF_SIZE,
-	};
+	struct iovec iov;
 	struct msghdr msg = {
 		.msg_name = &sa,
 		.msg_namelen = sizeof(sa),
@@ -342,18 +360,43 @@ mlx5_nl_recv(int nlsk_fd, uint32_t sn, int (*cb)(struct nlmsghdr *, void *arg),
 		/* One message at a time */
 		.msg_iovlen = 1,
 	};
+	void *buf = NULL;
 	int multipart = 0;
 	int ret = 0;
 
-	if (!buf) {
-		rte_errno = ENOMEM;
-		return -rte_errno;
-	}
 	do {
 		struct nlmsghdr *nh;
-		int recv_bytes = 0;
+		int recv_bytes;
 
 		do {
+			/* Query length of incoming message. */
+			iov.iov_base = NULL;
+			iov.iov_len = 0;
+			recv_bytes = recvmsg(nlsk_fd, &msg,
+					     MSG_PEEK | MSG_TRUNC);
+			if (recv_bytes < 0) {
+				rte_errno = errno;
+				ret = -rte_errno;
+				goto exit;
+			}
+			if (recv_bytes == 0) {
+				rte_errno = ENODATA;
+				ret = -rte_errno;
+				goto exit;
+			}
+			/* Allocate buffer to fetch the message. */
+			if (recv_bytes < MLX5_RECV_BUF_SIZE)
+				recv_bytes = MLX5_RECV_BUF_SIZE;
+			mlx5_free(buf);
+			buf = mlx5_malloc(0, recv_bytes, 0, SOCKET_ID_ANY);
+			if (!buf) {
+				rte_errno = ENOMEM;
+				ret = -rte_errno;
+				goto exit;
+			}
+			/* Fetch the message. */
+			iov.iov_base = buf;
+			iov.iov_len = recv_bytes;
 			recv_bytes = recvmsg(nlsk_fd, &msg, 0);
 			if (recv_bytes == -1) {
 				rte_errno = errno;
@@ -1188,6 +1231,7 @@ mlx5_nl_switch_info_cb(struct nlmsghdr *nh, void *arg)
 	size_t off = NLMSG_LENGTH(sizeof(struct ifinfomsg));
 	bool switch_id_set = false;
 	bool num_vf_set = false;
+	int len;
 
 	if (nh->nlmsg_type != RTM_NEWLINK)
 		goto error;
@@ -1203,7 +1247,24 @@ mlx5_nl_switch_info_cb(struct nlmsghdr *nh, void *arg)
 			num_vf_set = true;
 			break;
 		case IFLA_PHYS_PORT_NAME:
-			mlx5_translate_port_name((char *)payload, &info);
+			len = RTA_PAYLOAD(ra);
+			/* Some kernels do not pad attributes with zero. */
+			if (len > 0 && len < MLX5_PHYS_PORT_NAME_MAX) {
+				char name[MLX5_PHYS_PORT_NAME_MAX];
+
+				/*
+				 * We can't just patch the message with padding
+				 * zero - it might corrupt the following items
+				 * in the message, we have to copy the string
+				 * by attribute length and pad the copied one.
+				 */
+				memcpy(name, payload, len);
+				name[len] = 0;
+				mlx5_translate_port_name(name, &info);
+			} else {
+				info.name_type =
+					MLX5_PHYS_PORT_NAME_TYPE_UNKNOWN;
+			}
 			break;
 		case IFLA_PHYS_SWITCH_ID:
 			info.switch_id = 0;
diff --git a/drivers/common/mlx5/mlx5_common_mr.c b/drivers/common/mlx5/mlx5_common_mr.c
index 7c25541..d01f868 100644
--- a/drivers/common/mlx5/mlx5_common_mr.c
+++ b/drivers/common/mlx5/mlx5_common_mr.c
@@ -1061,6 +1061,95 @@ mlx5_create_mr_ext(void *pd, uintptr_t addr, size_t len, int socket_id,
 }
 
 /**
+ * Callback for memory free event. Iterate freed memsegs and check whether it
+ * belongs to an existing MR. If found, clear the bit from bitmap of MR. As a
+ * result, the MR would be fragmented. If it becomes empty, the MR will be freed
+ * later by mlx5_mr_garbage_collect(). Even if this callback is called from a
+ * secondary process, the garbage collector will be called in primary process
+ * as the secondary process can't call mlx5_mr_create().
+ *
+ * The global cache must be rebuilt if there's any change and this event has to
+ * be propagated to dataplane threads to flush the local caches.
+ *
+ * @param share_cache
+ *   Pointer to a global shared MR cache.
+ * @param ibdev_name
+ *   Name of ibv device.
+ * @param addr
+ *   Address of freed memory.
+ * @param len
+ *   Size of freed memory.
+ */
+void
+mlx5_free_mr_by_addr(struct mlx5_mr_share_cache *share_cache,
+		     const char *ibdev_name, const void *addr, size_t len)
+{
+	const struct rte_memseg_list *msl;
+	struct mlx5_mr *mr;
+	int ms_n;
+	int i;
+	int rebuild = 0;
+
+	DRV_LOG(DEBUG, "device %s free callback: addr=%p, len=%zu",
+		ibdev_name, addr, len);
+	msl = rte_mem_virt2memseg_list(addr);
+	/* addr and len must be page-aligned. */
+	MLX5_ASSERT((uintptr_t)addr ==
+		    RTE_ALIGN((uintptr_t)addr, msl->page_sz));
+	MLX5_ASSERT(len == RTE_ALIGN(len, msl->page_sz));
+	ms_n = len / msl->page_sz;
+	rte_rwlock_write_lock(&share_cache->rwlock);
+	/* Clear bits of freed memsegs from MR. */
+	for (i = 0; i < ms_n; ++i) {
+		const struct rte_memseg *ms;
+		struct mr_cache_entry entry;
+		uintptr_t start;
+		int ms_idx;
+		uint32_t pos;
+
+		/* Find MR having this memseg. */
+		start = (uintptr_t)addr + i * msl->page_sz;
+		mr = mlx5_mr_lookup_list(share_cache, &entry, start);
+		if (mr == NULL)
+			continue;
+		MLX5_ASSERT(mr->msl); /* Can't be external memory. */
+		ms = rte_mem_virt2memseg((void *)start, msl);
+		MLX5_ASSERT(ms != NULL);
+		MLX5_ASSERT(msl->page_sz == ms->hugepage_sz);
+		ms_idx = rte_fbarray_find_idx(&msl->memseg_arr, ms);
+		pos = ms_idx - mr->ms_base_idx;
+		MLX5_ASSERT(rte_bitmap_get(mr->ms_bmp, pos));
+		MLX5_ASSERT(pos < mr->ms_bmp_n);
+		DRV_LOG(DEBUG, "device %s MR(%p): clear bitmap[%u] for addr %p",
+			ibdev_name, (void *)mr, pos, (void *)start);
+		rte_bitmap_clear(mr->ms_bmp, pos);
+		if (--mr->ms_n == 0) {
+			LIST_REMOVE(mr, mr);
+			LIST_INSERT_HEAD(&share_cache->mr_free_list, mr, mr);
+			DRV_LOG(DEBUG, "device %s remove MR(%p) from list",
+				ibdev_name, (void *)mr);
+		}
+		/*
+		 * MR is fragmented or will be freed. the global cache must be
+		 * rebuilt.
+		 */
+		rebuild = 1;
+	}
+	if (rebuild) {
+		mlx5_mr_rebuild_cache(share_cache);
+		/*
+		 * No explicit wmb is needed after updating dev_gen due to
+		 * store-release ordering in unlock that provides the
+		 * implicit barrier at the software visible level.
+		 */
+		++share_cache->dev_gen;
+		DRV_LOG(DEBUG, "broadcasting local cache flush, gen=%d",
+			share_cache->dev_gen);
+	}
+	rte_rwlock_write_unlock(&share_cache->rwlock);
+}
+
+/**
  * Dump all the created MRs and the global cache entries.
  *
  * @param sh
diff --git a/drivers/common/mlx5/mlx5_common_mr.h b/drivers/common/mlx5/mlx5_common_mr.h
index da0a0f0..09d39dd 100644
--- a/drivers/common/mlx5/mlx5_common_mr.h
+++ b/drivers/common/mlx5/mlx5_common_mr.h
@@ -143,6 +143,9 @@ void mlx5_mr_rebuild_cache(struct mlx5_mr_share_cache *share_cache);
 __rte_internal
 void mlx5_mr_flush_local_cache(struct mlx5_mr_ctrl *mr_ctrl);
 __rte_internal
+void mlx5_free_mr_by_addr(struct mlx5_mr_share_cache *share_cache,
+			  const char *ibdev_name, const void *addr, size_t len);
+__rte_internal
 int
 mlx5_mr_insert_cache(struct mlx5_mr_share_cache *share_cache,
 		     struct mlx5_mr *mr);
diff --git a/drivers/common/mlx5/version.map b/drivers/common/mlx5/version.map
index fd6019b..02aedbc 100644
--- a/drivers/common/mlx5/version.map
+++ b/drivers/common/mlx5/version.map
@@ -69,6 +69,7 @@ INTERNAL {
 	mlx5_mr_create_primary;
 	mlx5_mr_flush_local_cache;
 	mlx5_mr_free;
+	mlx5_free_mr_by_addr;
 
 	mlx5_nl_allmulti;
 	mlx5_nl_devlink_family_id_get;
diff --git a/drivers/common/octeontx2/otx2_dev.h b/drivers/common/octeontx2/otx2_dev.h
index cd4fe51..9d8dcca 100644
--- a/drivers/common/octeontx2/otx2_dev.h
+++ b/drivers/common/octeontx2/otx2_dev.h
@@ -55,6 +55,9 @@
 	 (RVU_PCI_REV_MINOR(otx2_dev_revid(dev)) == 0x0) &&	\
 	 (RVU_PCI_REV_MIDR_ID(otx2_dev_revid(dev)) == 0x0))
 
+#define otx2_dev_is_98xx(dev)                                   \
+	 (RVU_PCI_REV_MIDR_ID(otx2_dev_revid(dev)) == 0x3)
+
 struct otx2_dev;
 
 /* Link status callback */
diff --git a/drivers/common/sfc_efx/meson.build b/drivers/common/sfc_efx/meson.build
index 6cb9f07..1e17f1f 100644
--- a/drivers/common/sfc_efx/meson.build
+++ b/drivers/common/sfc_efx/meson.build
@@ -5,7 +5,7 @@
 # This software was jointly developed between OKTET Labs (under contract
 # for Solarflare) and Solarflare Communications, Inc.
 
-if (arch_subdir != 'x86' or not dpdk_conf.get('RTE_ARCH_64')) and (arch_subdir != 'arm' or not host_machine.cpu_family().startswith('aarch64'))
+if (arch_subdir != 'x86' and arch_subdir != 'arm') or (not dpdk_conf.get('RTE_ARCH_64'))
 	build = false
 	reason = 'only supported on x86_64 and aarch64'
 endif
diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
index 0de5120..cc4e085 100644
--- a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
+++ b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
@@ -842,8 +842,14 @@ aesni_gcm_create(const char *name,
 		init_mb_mgr_avx2(mb_mgr);
 		break;
 	case RTE_AESNI_GCM_AVX512:
-		dev->feature_flags |= RTE_CRYPTODEV_FF_CPU_AVX512;
-		init_mb_mgr_avx512(mb_mgr);
+		if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_VAES)) {
+			dev->feature_flags |= RTE_CRYPTODEV_FF_CPU_AVX512;
+			init_mb_mgr_avx512(mb_mgr);
+		} else {
+			dev->feature_flags |= RTE_CRYPTODEV_FF_CPU_AVX2;
+			init_mb_mgr_avx2(mb_mgr);
+			vector_mode = RTE_AESNI_GCM_AVX2;
+		}
 		break;
 	default:
 		AESNI_GCM_LOG(ERR, "Unsupported vector mode %u\n", vector_mode);
diff --git a/drivers/crypto/mvsam/rte_mrvl_pmd.c b/drivers/crypto/mvsam/rte_mrvl_pmd.c
index bec51c9..5712042 100644
--- a/drivers/crypto/mvsam/rte_mrvl_pmd.c
+++ b/drivers/crypto/mvsam/rte_mrvl_pmd.c
@@ -360,6 +360,14 @@ mrvl_crypto_set_aead_session_parameters(struct mrvl_crypto_session *sess,
 	sess->sam_sess_params.cipher_mode =
 		aead_map[aead_xform->aead.algo].cipher_mode;
 
+	if (sess->sam_sess_params.cipher_mode == SAM_CIPHER_GCM) {
+		/* IV must include nonce for all counter modes */
+		sess->cipher_iv_offset = aead_xform->cipher.iv.offset;
+
+		/* Set order of authentication then encryption to 0 in GCM */
+		sess->sam_sess_params.u.basic.auth_then_encrypt = 0;
+	}
+
 	/* Assume IV will be passed together with data. */
 	sess->sam_sess_params.cipher_iv = NULL;
 
@@ -916,14 +924,14 @@ mrvl_pmd_parse_input_args(struct mrvl_pmd_init_params *params,
 		ret = rte_kvargs_process(kvlist,
 					 RTE_CRYPTODEV_PMD_NAME_ARG,
 					 &parse_name_arg,
-					 &params->common);
+					 &params->common.name);
 		if (ret < 0)
 			goto free_kvlist;
 
 		ret = rte_kvargs_process(kvlist,
 					 MRVL_PMD_MAX_NB_SESS_ARG,
 					 &parse_integer_arg,
-					 params);
+					 &params->max_nb_sessions);
 		if (ret < 0)
 			goto free_kvlist;
 
diff --git a/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c b/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
index c61bdca..4eb7ec9 100644
--- a/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
+++ b/drivers/crypto/mvsam/rte_mrvl_pmd_ops.c
@@ -111,7 +111,7 @@ static const struct rte_cryptodev_capabilities
 					.increment = 1
 				},
 				.digest_size = {
-					.min = 28,
+					.min = 12,
 					.max = 28,
 					.increment = 0
 				},
@@ -232,7 +232,7 @@ static const struct rte_cryptodev_capabilities
 				},
 				.digest_size = {
 					.min = 12,
-					.max = 48,
+					.max = 64,
 					.increment = 4
 				},
 			}, }
@@ -252,7 +252,7 @@ static const struct rte_cryptodev_capabilities
 				},
 				.digest_size = {
 					.min = 12,
-					.max = 48,
+					.max = 64,
 					.increment = 0
 				},
 			}, }
@@ -336,9 +336,9 @@ static const struct rte_cryptodev_capabilities
 					.increment = 0
 				},
 				.aad_size = {
-					.min = 8,
-					.max = 12,
-					.increment = 4
+					.min = 0,
+					.max = 64,
+					.increment = 1
 				},
 				.iv_size = {
 					.min = 12,
@@ -793,7 +793,7 @@ mrvl_crypto_pmd_sym_session_clear(struct rte_cryptodev *dev,
 			MRVL_LOG(ERR, "Error while destroying session!");
 		}
 
-		memset(sess, 0, sizeof(struct mrvl_crypto_session));
+		memset(mrvl_sess, 0, sizeof(struct mrvl_crypto_session));
 		struct rte_mempool *sess_mp = rte_mempool_from_obj(sess_priv);
 		set_sym_session_private_data(sess, index, NULL);
 		rte_mempool_put(sess_mp, sess_priv);
diff --git a/drivers/crypto/octeontx/otx_cryptodev.c b/drivers/crypto/octeontx/otx_cryptodev.c
index 5ce1cf8..36cedb3 100644
--- a/drivers/crypto/octeontx/otx_cryptodev.c
+++ b/drivers/crypto/octeontx/otx_cryptodev.c
@@ -71,6 +71,7 @@ otx_cpt_pci_remove(struct rte_pci_device *pci_dev)
 {
 	struct rte_cryptodev *cryptodev;
 	char name[RTE_CRYPTODEV_NAME_MAX_LEN];
+	void *dev_priv;
 
 	if (pci_dev == NULL)
 		return -EINVAL;
@@ -84,11 +85,13 @@ otx_cpt_pci_remove(struct rte_pci_device *pci_dev)
 	if (pci_dev->driver == NULL)
 		return -ENODEV;
 
+	dev_priv = cryptodev->data->dev_private;
+
 	/* free crypto device */
 	rte_cryptodev_pmd_release_device(cryptodev);
 
 	if (rte_eal_process_type() == RTE_PROC_PRIMARY)
-		rte_free(cryptodev->data->dev_private);
+		rte_free(dev_priv);
 
 	cryptodev->device->driver = NULL;
 	cryptodev->device = NULL;
diff --git a/drivers/crypto/octeontx2/otx2_ipsec_po_ops.h b/drivers/crypto/octeontx2/otx2_ipsec_po_ops.h
index bc702d5..ee30131 100644
--- a/drivers/crypto/octeontx2/otx2_ipsec_po_ops.h
+++ b/drivers/crypto/octeontx2/otx2_ipsec_po_ops.h
@@ -61,16 +61,12 @@ process_outb_sa(struct rte_crypto_op *cop,
 	uint32_t dlen, rlen, extend_head, extend_tail;
 	struct rte_crypto_sym_op *sym_op = cop->sym;
 	struct rte_mbuf *m_src = sym_op->m_src;
-	struct otx2_ipsec_po_sa_ctl *ctl_wrd;
 	struct cpt_request_info *req = NULL;
 	struct otx2_ipsec_po_out_hdr *hdr;
-	struct otx2_ipsec_po_out_sa *sa;
 	int hdr_len, mdata_len, ret = 0;
 	vq_cmd_word0_t word0;
 	char *mdata, *data;
 
-	sa = &sess->out_sa;
-	ctl_wrd = &sa->ctl;
 	hdr_len = sizeof(*hdr);
 
 	dlen = rte_pktmbuf_pkt_len(m_src) + hdr_len;
@@ -107,14 +103,8 @@ process_outb_sa(struct rte_crypto_op *cop,
 	hdr = (struct otx2_ipsec_po_out_hdr *)rte_pktmbuf_adj(m_src,
 							RTE_ETHER_HDR_LEN);
 
-	if (ctl_wrd->enc_type == OTX2_IPSEC_FP_SA_ENC_AES_GCM) {
-		memcpy(&hdr->iv[0], &sa->iv.gcm.nonce, 4);
-		memcpy(&hdr->iv[4], rte_crypto_op_ctod_offset(cop, uint8_t *,
-			sess->iv_offset), sess->iv_length);
-	} else if (ctl_wrd->auth_type == OTX2_IPSEC_FP_SA_ENC_AES_CBC) {
-		memcpy(&hdr->iv[0], rte_crypto_op_ctod_offset(cop, uint8_t *,
-			sess->iv_offset), sess->iv_length);
-	}
+	memcpy(&hdr->iv[0], rte_crypto_op_ctod_offset(cop, uint8_t *,
+		sess->iv_offset), sess->iv_length);
 
 	/* Prepare CPT instruction */
 	word0.u64 = sess->ucmd_w0;
diff --git a/drivers/crypto/octeontx2/otx2_security.h b/drivers/crypto/octeontx2/otx2_security.h
index 33d3b15..7087ea3 100644
--- a/drivers/crypto/octeontx2/otx2_security.h
+++ b/drivers/crypto/octeontx2/otx2_security.h
@@ -19,14 +19,16 @@
 #define OTX2_SEC_AES_GCM_ROUNDUP_BYTE_LEN	4
 #define OTX2_SEC_AES_CBC_ROUNDUP_BYTE_LEN	16
 
-union otx2_sec_session_ipsec {
-	struct otx2_sec_session_ipsec_ip ip;
-	struct otx2_sec_session_ipsec_lp lp;
+struct otx2_sec_session_ipsec {
+	union {
+		struct otx2_sec_session_ipsec_ip ip;
+		struct otx2_sec_session_ipsec_lp lp;
+	};
 	enum rte_security_ipsec_sa_direction dir;
 };
 
 struct otx2_sec_session {
-	union otx2_sec_session_ipsec ipsec;
+	struct otx2_sec_session_ipsec ipsec;
 	void *userdata;
 	/**< Userdata registered by the application */
 } __rte_cache_aligned;
diff --git a/drivers/crypto/qat/qat_asym_pmd.c b/drivers/crypto/qat/qat_asym_pmd.c
index a2c8aca..ef47e28 100644
--- a/drivers/crypto/qat/qat_asym_pmd.c
+++ b/drivers/crypto/qat/qat_asym_pmd.c
@@ -251,6 +251,10 @@ qat_asym_dev_create(struct qat_pci_device *qat_pci_dev,
 	struct rte_cryptodev *cryptodev;
 	struct qat_asym_dev_private *internals;
 
+	if (qat_pci_dev->qat_dev_gen == QAT_GEN3) {
+		QAT_LOG(ERR, "Asymmetric crypto PMD not supported on QAT c4xxx");
+		return -EFAULT;
+	}
 	snprintf(name, RTE_CRYPTODEV_NAME_MAX_LEN, "%s_%s",
 			qat_pci_dev->name, "asym");
 	QAT_LOG(DEBUG, "Creating QAT ASYM device %s\n", name);
diff --git a/drivers/crypto/qat/qat_sym_session.c b/drivers/crypto/qat/qat_sym_session.c
index 23d059b..6e2193f 100644
--- a/drivers/crypto/qat/qat_sym_session.c
+++ b/drivers/crypto/qat/qat_sym_session.c
@@ -1190,6 +1190,9 @@ static int partial_hash_compute(enum icp_qat_hw_auth_algo hash_alg,
 	uint64_t *hash_state_out_be64;
 	int i;
 
+	/* Initialize to avoid gcc warning */
+	memset(digest, 0, sizeof(digest));
+
 	digest_size = qat_hash_get_digest_size(hash_alg);
 	if (digest_size <= 0)
 		return -EFAULT;
diff --git a/drivers/event/octeontx2/otx2_worker.h b/drivers/event/octeontx2/otx2_worker.h
index 0a7d667..89dc79d 100644
--- a/drivers/event/octeontx2/otx2_worker.h
+++ b/drivers/event/octeontx2/otx2_worker.h
@@ -272,7 +272,7 @@ otx2_ssogws_prepare_pkt(const struct otx2_eth_txq *txq, struct rte_mbuf *m,
 			uint64_t *cmd, const uint32_t flags)
 {
 	otx2_lmt_mov(cmd, txq->cmd, otx2_nix_tx_ext_subs(flags));
-	otx2_nix_xmit_prepare(m, cmd, flags);
+	otx2_nix_xmit_prepare(m, cmd, flags, txq->lso_tun_fmt);
 }
 
 static __rte_always_inline uint16_t
diff --git a/drivers/mempool/octeontx2/otx2_mempool_ops.c b/drivers/mempool/octeontx2/otx2_mempool_ops.c
index 9ff71bc..d827fd8 100644
--- a/drivers/mempool/octeontx2/otx2_mempool_ops.c
+++ b/drivers/mempool/octeontx2/otx2_mempool_ops.c
@@ -611,7 +611,8 @@ npa_lf_aura_pool_pair_alloc(struct otx2_npa_lf *lf, const uint32_t block_size,
 	/* Update aura fields */
 	aura->pool_addr = pool_id;/* AF will translate to associated poolctx */
 	aura->ena = 1;
-	aura->shift = __builtin_clz(block_count) - 8;
+	aura->shift = rte_log2_u32(block_count);
+	aura->shift = aura->shift < 8 ? 0 : aura->shift - 8;
 	aura->limit = block_count;
 	aura->pool_caching = 1;
 	aura->err_int_ena = BIT(NPA_AURA_ERR_INT_AURA_ADD_OVER);
@@ -626,7 +627,8 @@ npa_lf_aura_pool_pair_alloc(struct otx2_npa_lf *lf, const uint32_t block_size,
 	pool->ena = 1;
 	pool->buf_size = block_size / OTX2_ALIGN;
 	pool->stack_max_pages = stack_size;
-	pool->shift = __builtin_clz(block_count) - 8;
+	pool->shift = rte_log2_u32(block_count);
+	pool->shift = pool->shift < 8 ? 0 : pool->shift - 8;
 	pool->ptr_start = 0;
 	pool->ptr_end = ~0;
 	pool->stack_caching = 1;
diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 0fb195b..f854098 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -605,6 +605,49 @@ struct bnxt_flow_stat_info {
 	struct bnxt_ctx_mem_buf_info tx_fc_out_tbl;
 };
 
+struct bnxt_ring_stats {
+	/* Number of transmitted unicast packets */
+	uint64_t	tx_ucast_pkts;
+	/* Number of transmitted multicast packets */
+	uint64_t	tx_mcast_pkts;
+	/* Number of transmitted broadcast packets */
+	uint64_t	tx_bcast_pkts;
+	/* Number of packets discarded in transmit path */
+	uint64_t	tx_discard_pkts;
+	/* Number of packets in transmit path with error */
+	uint64_t	tx_error_pkts;
+	/* Number of transmitted bytes for unicast traffic */
+	uint64_t	tx_ucast_bytes;
+	/* Number of transmitted bytes for multicast traffic */
+	uint64_t	tx_mcast_bytes;
+	/* Number of transmitted bytes for broadcast traffic */
+	uint64_t	tx_bcast_bytes;
+	/* Number of received unicast packets */
+	uint64_t	rx_ucast_pkts;
+	/* Number of received multicast packets */
+	uint64_t	rx_mcast_pkts;
+	/* Number of received broadcast packets */
+	uint64_t	rx_bcast_pkts;
+	/* Number of packets discarded in receive path */
+	uint64_t	rx_discard_pkts;
+	/* Number of packets in receive path with errors */
+	uint64_t	rx_error_pkts;
+	/* Number of received bytes for unicast traffic */
+	uint64_t	rx_ucast_bytes;
+	/* Number of received bytes for multicast traffic */
+	uint64_t	rx_mcast_bytes;
+	/* Number of received bytes for broadcast traffic */
+	uint64_t	rx_bcast_bytes;
+	/* Number of aggregated unicast packets */
+	uint64_t	rx_agg_pkts;
+	/* Number of aggregated unicast bytes */
+	uint64_t	rx_agg_bytes;
+	/* Number of aggregation events */
+	uint64_t	rx_agg_events;
+	/* Number of aborted aggregations */
+	uint64_t	rx_agg_aborts;
+};
+
 struct bnxt {
 	void				*bar0;
 
@@ -808,6 +851,8 @@ struct bnxt {
 	uint8_t			flow_xstat;
 	uint16_t		max_num_kflows;
 	uint16_t		tx_cfa_action;
+	struct bnxt_ring_stats	*prev_rx_ring_stats;
+	struct bnxt_ring_stats	*prev_tx_ring_stats;
 };
 
 static
diff --git a/drivers/net/bnxt/bnxt_cpr.c b/drivers/net/bnxt/bnxt_cpr.c
index ee96ae8..7bfda01 100644
--- a/drivers/net/bnxt/bnxt_cpr.c
+++ b/drivers/net/bnxt/bnxt_cpr.c
@@ -109,6 +109,8 @@ void bnxt_handle_async_event(struct bnxt *bp,
 	case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE:
 		/* FALLTHROUGH */
 		bnxt_link_update_op(bp->eth_dev, 0);
+		rte_eth_dev_callback_process(bp->eth_dev,
+			RTE_ETH_EVENT_INTR_LSC, NULL);
 		break;
 	case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD:
 		PMD_DRV_LOG(INFO, "Async event: PF driver unloaded\n");
diff --git a/drivers/net/bnxt/bnxt_cpr.h b/drivers/net/bnxt/bnxt_cpr.h
index c769bde..fedfb47 100644
--- a/drivers/net/bnxt/bnxt_cpr.h
+++ b/drivers/net/bnxt/bnxt_cpr.h
@@ -8,13 +8,10 @@
 #include <stdbool.h>
 
 #include <rte_io.h>
+#include "hsi_struct_def_dpdk.h"
 
 struct bnxt_db_info;
 
-#define CMP_VALID(cmp, raw_cons, ring)					\
-	(!!(rte_le_to_cpu_32(((struct cmpl_base *)(cmp))->info3_v) &	\
-	    CMPL_BASE_V) == !((raw_cons) & ((ring)->ring_size)))
-
 #define CMPL_VALID(cmp, v)						\
 	(!!(rte_le_to_cpu_32(((struct cmpl_base *)(cmp))->info3_v) &	\
 	    CMPL_BASE_V) == !(v))
@@ -131,4 +128,35 @@ bool bnxt_is_recovery_enabled(struct bnxt *bp);
 bool bnxt_is_master_func(struct bnxt *bp);
 
 void bnxt_stop_rxtx(struct bnxt *bp);
+
+/**
+ * Check validity of a completion ring entry. If the entry is valid, include a
+ * C11 __ATOMIC_ACQUIRE fence to ensure that subsequent loads of fields in the
+ * completion are not hoisted by the compiler or by the CPU to come before the
+ * loading of the "valid" field.
+ *
+ * Note: the caller must not access any fields in the specified completion
+ * entry prior to calling this function.
+ *
+ * @param cmpl
+ *   Pointer to an entry in the completion ring.
+ * @param raw_cons
+ *   Raw consumer index of entry in completion ring.
+ * @param ring_size
+ *   Size of completion ring.
+ */
+static __rte_always_inline bool
+bnxt_cpr_cmp_valid(const void *cmpl, uint32_t raw_cons, uint32_t ring_size)
+{
+	const struct cmpl_base *c = cmpl;
+	bool expected, valid;
+
+	expected = !(raw_cons & ring_size);
+	valid = !!(rte_le_to_cpu_32(c->info3_v) & CMPL_BASE_V);
+	if (valid == expected) {
+		rte_atomic_thread_fence(__ATOMIC_ACQUIRE);
+		return true;
+	}
+	return false;
+}
 #endif
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index 1498f4a..8afe72b 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -583,13 +583,14 @@ static int bnxt_register_fc_ctx_mem(struct bnxt *bp)
 	return rc;
 }
 
-static int bnxt_alloc_ctx_mem_buf(char *type, size_t size,
+static int bnxt_alloc_ctx_mem_buf(struct bnxt *bp, char *type, size_t size,
 				  struct bnxt_ctx_mem_buf_info *ctx)
 {
 	if (!ctx)
 		return -EINVAL;
 
-	ctx->va = rte_zmalloc(type, size, 0);
+	ctx->va = rte_zmalloc_socket(type, size, 0,
+				     bp->eth_dev->device->numa_node);
 	if (ctx->va == NULL)
 		return -ENOMEM;
 	rte_mem_lock_page(ctx->va);
@@ -613,7 +614,7 @@ static int bnxt_init_fc_ctx_mem(struct bnxt *bp)
 	sprintf(type, "bnxt_rx_fc_in_" PCI_PRI_FMT, pdev->addr.domain,
 		pdev->addr.bus, pdev->addr.devid, pdev->addr.function);
 	/* 4 bytes for each counter-id */
-	rc = bnxt_alloc_ctx_mem_buf(type,
+	rc = bnxt_alloc_ctx_mem_buf(bp, type,
 				    max_fc * 4,
 				    &bp->flow_stat->rx_fc_in_tbl);
 	if (rc)
@@ -622,7 +623,7 @@ static int bnxt_init_fc_ctx_mem(struct bnxt *bp)
 	sprintf(type, "bnxt_rx_fc_out_" PCI_PRI_FMT, pdev->addr.domain,
 		pdev->addr.bus, pdev->addr.devid, pdev->addr.function);
 	/* 16 bytes for each counter - 8 bytes pkt_count, 8 bytes byte_count */
-	rc = bnxt_alloc_ctx_mem_buf(type,
+	rc = bnxt_alloc_ctx_mem_buf(bp, type,
 				    max_fc * 16,
 				    &bp->flow_stat->rx_fc_out_tbl);
 	if (rc)
@@ -631,7 +632,7 @@ static int bnxt_init_fc_ctx_mem(struct bnxt *bp)
 	sprintf(type, "bnxt_tx_fc_in_" PCI_PRI_FMT, pdev->addr.domain,
 		pdev->addr.bus, pdev->addr.devid, pdev->addr.function);
 	/* 4 bytes for each counter-id */
-	rc = bnxt_alloc_ctx_mem_buf(type,
+	rc = bnxt_alloc_ctx_mem_buf(bp, type,
 				    max_fc * 4,
 				    &bp->flow_stat->tx_fc_in_tbl);
 	if (rc)
@@ -640,7 +641,7 @@ static int bnxt_init_fc_ctx_mem(struct bnxt *bp)
 	sprintf(type, "bnxt_tx_fc_out_" PCI_PRI_FMT, pdev->addr.domain,
 		pdev->addr.bus, pdev->addr.devid, pdev->addr.function);
 	/* 16 bytes for each counter - 8 bytes pkt_count, 8 bytes byte_count */
-	rc = bnxt_alloc_ctx_mem_buf(type,
+	rc = bnxt_alloc_ctx_mem_buf(bp, type,
 				    max_fc * 16,
 				    &bp->flow_stat->tx_fc_out_tbl);
 	if (rc)
@@ -696,6 +697,38 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
 	return rc;
 }
 
+static void bnxt_free_prev_ring_stats(struct bnxt *bp)
+{
+	rte_free(bp->prev_rx_ring_stats);
+	rte_free(bp->prev_tx_ring_stats);
+
+	bp->prev_rx_ring_stats = NULL;
+	bp->prev_tx_ring_stats = NULL;
+}
+
+static int bnxt_alloc_prev_ring_stats(struct bnxt *bp)
+{
+	bp->prev_rx_ring_stats =  rte_zmalloc("bnxt_prev_rx_ring_stats",
+					      sizeof(struct bnxt_ring_stats) *
+					      bp->rx_cp_nr_rings,
+					      0);
+	if (bp->prev_rx_ring_stats == NULL)
+		return -ENOMEM;
+
+	bp->prev_tx_ring_stats = rte_zmalloc("bnxt_prev_tx_ring_stats",
+					     sizeof(struct bnxt_ring_stats) *
+					     bp->tx_cp_nr_rings,
+					     0);
+	if (bp->prev_tx_ring_stats == NULL)
+		goto error;
+
+	return 0;
+
+error:
+	bnxt_free_prev_ring_stats(bp);
+	return -ENOMEM;
+}
+
 static int bnxt_start_nic(struct bnxt *bp)
 {
 	struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(bp->eth_dev);
@@ -933,7 +966,7 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev,
 	dev_info->max_rx_queues = max_rx_rings;
 	dev_info->max_tx_queues = max_rx_rings;
 	dev_info->reta_size = bnxt_rss_hash_tbl_size(bp);
-	dev_info->hash_key_size = 40;
+	dev_info->hash_key_size = HW_HASH_KEY_SIZE;
 	max_vnics = bp->max_vnics;
 
 	/* MTU specifics */
@@ -954,7 +987,6 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev,
 
 	dev_info->speed_capa = bnxt_get_speed_capabilities(bp);
 
-	/* *INDENT-OFF* */
 	dev_info->default_rxconf = (struct rte_eth_rxconf) {
 		.rx_thresh = {
 			.pthresh = 8,
@@ -976,7 +1008,6 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev,
 	};
 	eth_dev->data->dev_conf.intr_conf.lsc = 1;
 
-	eth_dev->data->dev_conf.intr_conf.rxq = 1;
 	dev_info->rx_desc_lim.nb_min = BNXT_MIN_RING_DESC;
 	dev_info->rx_desc_lim.nb_max = BNXT_MAX_RX_RING_DESC;
 	dev_info->tx_desc_lim.nb_min = BNXT_MIN_RING_DESC;
@@ -990,8 +1021,6 @@ static int bnxt_dev_info_get_op(struct rte_eth_dev *eth_dev,
 				    BNXT_SWITCH_PORT_ID_TRUSTED_VF;
 	}
 
-	/* *INDENT-ON* */
-
 	/*
 	 * TODO: default_rxconf, default_txconf, rx_desc_lim, and tx_desc_lim
 	 *       need further investigation.
@@ -1435,6 +1464,7 @@ static int bnxt_dev_stop_op(struct rte_eth_dev *eth_dev)
 	bnxt_shutdown_nic(bp);
 	bnxt_hwrm_if_change(bp, false);
 
+	bnxt_free_prev_ring_stats(bp);
 	rte_free(bp->mark_table);
 	bp->mark_table = NULL;
 
@@ -1489,6 +1519,10 @@ static int bnxt_dev_start_op(struct rte_eth_dev *eth_dev)
 	if (rc)
 		goto error;
 
+	rc = bnxt_alloc_prev_ring_stats(bp);
+	if (rc)
+		goto error;
+
 	eth_dev->data->dev_started = 1;
 
 	bnxt_link_update_op(eth_dev, 1);
@@ -1732,11 +1766,6 @@ out:
 	if (new.link_status != eth_dev->data->dev_link.link_status ||
 	new.link_speed != eth_dev->data->dev_link.link_speed) {
 		rte_eth_linkstatus_set(eth_dev, &new);
-
-		rte_eth_dev_callback_process(eth_dev,
-					     RTE_ETH_EVENT_INTR_LSC,
-					     NULL);
-
 		bnxt_print_link_info(eth_dev);
 	}
 
@@ -1962,7 +1991,6 @@ static int bnxt_reta_query_op(struct rte_eth_dev *eth_dev,
 	if (rc)
 		return rc;
 
-	/* Retrieve from the default VNIC */
 	if (!vnic)
 		return -EINVAL;
 	if (!vnic->rss_table)
@@ -2044,7 +2072,8 @@ static int bnxt_rss_hash_update_op(struct rte_eth_dev *eth_dev,
 
 	if (rss_conf->rss_key_len != HW_HASH_KEY_SIZE) {
 		PMD_DRV_LOG(ERR,
-			    "Invalid hashkey length, should be 16 bytes\n");
+			    "Invalid hashkey length, should be %d bytes\n",
+			    HW_HASH_KEY_SIZE);
 		return -EINVAL;
 	}
 	memcpy(vnic->rss_hash_key, rss_conf->rss_key, rss_conf->rss_key_len);
@@ -3007,7 +3036,7 @@ bnxt_rx_queue_count_op(struct rte_eth_dev *dev, uint16_t rx_queue_id)
 {
 	struct bnxt *bp = (struct bnxt *)dev->data->dev_private;
 	struct bnxt_cp_ring_info *cpr;
-	uint32_t desc = 0, raw_cons;
+	uint32_t desc = 0, raw_cons, cp_ring_size;
 	struct bnxt_rx_queue *rxq;
 	struct rx_pkt_cmpl *rxcmp;
 	int rc;
@@ -3019,6 +3048,7 @@ bnxt_rx_queue_count_op(struct rte_eth_dev *dev, uint16_t rx_queue_id)
 	rxq = dev->data->rx_queues[rx_queue_id];
 	cpr = rxq->cp_ring;
 	raw_cons = cpr->cp_raw_cons;
+	cp_ring_size = cpr->cp_ring_struct->ring_size;
 
 	while (1) {
 		uint32_t agg_cnt, cons, cmpl_type;
@@ -3026,7 +3056,7 @@ bnxt_rx_queue_count_op(struct rte_eth_dev *dev, uint16_t rx_queue_id)
 		cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
 		rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
 
-		if (!CMP_VALID(rxcmp, raw_cons, cpr->cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(rxcmp, raw_cons, cp_ring_size))
 			break;
 
 		cmpl_type = CMP_TYPE(rxcmp);
@@ -3070,7 +3100,7 @@ bnxt_rx_descriptor_status_op(void *rx_queue, uint16_t offset)
 	struct bnxt_rx_queue *rxq = rx_queue;
 	struct bnxt_cp_ring_info *cpr;
 	struct bnxt_rx_ring_info *rxr;
-	uint32_t desc, raw_cons;
+	uint32_t desc, raw_cons, cp_ring_size;
 	struct bnxt *bp = rxq->bp;
 	struct rx_pkt_cmpl *rxcmp;
 	int rc;
@@ -3084,6 +3114,7 @@ bnxt_rx_descriptor_status_op(void *rx_queue, uint16_t offset)
 
 	rxr = rxq->rx_ring;
 	cpr = rxq->cp_ring;
+	cp_ring_size = cpr->cp_ring_struct->ring_size;
 
 	/*
 	 * For the vector receive case, the completion at the requested
@@ -3100,7 +3131,7 @@ bnxt_rx_descriptor_status_op(void *rx_queue, uint16_t offset)
 		cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
 		rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
 
-		if (CMP_VALID(rxcmp, raw_cons, cpr->cp_ring_struct))
+		if (bnxt_cpr_cmp_valid(rxcmp, raw_cons, cp_ring_size))
 			return RTE_ETH_RX_DESC_DONE;
 
 		/* Check whether rx desc has an mbuf attached. */
@@ -3126,7 +3157,7 @@ bnxt_rx_descriptor_status_op(void *rx_queue, uint16_t offset)
 		cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
 		rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
 
-		if (!CMP_VALID(rxcmp, raw_cons, cpr->cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(rxcmp, raw_cons, cp_ring_size))
 			break;
 
 		cmpl_type = CMP_TYPE(rxcmp);
@@ -3178,41 +3209,47 @@ static int
 bnxt_tx_descriptor_status_op(void *tx_queue, uint16_t offset)
 {
 	struct bnxt_tx_queue *txq = (struct bnxt_tx_queue *)tx_queue;
-	struct bnxt_tx_ring_info *txr;
-	struct bnxt_cp_ring_info *cpr;
-	struct bnxt_sw_tx_bd *tx_buf;
-	struct tx_pkt_cmpl *txcmp;
-	uint32_t cons, cp_cons;
+	struct bnxt_cp_ring_info *cpr = txq->cp_ring;
+	uint32_t ring_mask, raw_cons, nb_tx_pkts = 0;
+	struct cmpl_base *cp_desc_ring;
 	int rc;
 
-	if (!txq)
-		return -EINVAL;
-
 	rc = is_bnxt_in_error(txq->bp);
 	if (rc)
 		return rc;
 
-	cpr = txq->cp_ring;
-	txr = txq->tx_ring;
-
 	if (offset >= txq->nb_tx_desc)
 		return -EINVAL;
 
-	cons = RING_CMP(cpr->cp_ring_struct, offset);
-	txcmp = (struct tx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
-	cp_cons = cpr->cp_raw_cons;
+	/* Return "desc done" if descriptor is available for use. */
+	if (bnxt_tx_bds_in_hw(txq) <= offset)
+		return RTE_ETH_TX_DESC_DONE;
 
-	if (cons > cp_cons) {
-		if (CMPL_VALID(txcmp, cpr->valid))
-			return RTE_ETH_TX_DESC_UNAVAIL;
-	} else {
-		if (CMPL_VALID(txcmp, !cpr->valid))
-			return RTE_ETH_TX_DESC_UNAVAIL;
+	raw_cons = cpr->cp_raw_cons;
+	cp_desc_ring = cpr->cp_desc_ring;
+	ring_mask = cpr->cp_ring_struct->ring_mask;
+
+	/* Check to see if hw has posted a completion for the descriptor. */
+	while (1) {
+		struct tx_cmpl *txcmp;
+		uint32_t cons;
+
+		cons = RING_CMPL(ring_mask, raw_cons);
+		txcmp = (struct tx_cmpl *)&cp_desc_ring[cons];
+
+		if (!bnxt_cpr_cmp_valid(txcmp, raw_cons, ring_mask + 1))
+			break;
+
+		if (CMP_TYPE(txcmp) == TX_CMPL_TYPE_TX_L2)
+			nb_tx_pkts += rte_le_to_cpu_32(txcmp->opaque);
+
+		if (nb_tx_pkts > offset)
+			return RTE_ETH_TX_DESC_DONE;
+
+		raw_cons = NEXT_RAW_CMP(raw_cons);
 	}
-	tx_buf = &txr->tx_buf_ring[cons];
-	if (tx_buf->mbuf == NULL)
-		return RTE_ETH_TX_DESC_DONE;
 
+	/* Descriptor is pending transmit, not yet completed by hardware. */
 	return RTE_ETH_TX_DESC_FULL;
 }
 
@@ -4096,6 +4133,10 @@ err_start:
 err:
 	bp->flags |= BNXT_FLAG_FATAL_ERROR;
 	bnxt_uninit_resources(bp, false);
+	if (bp->eth_dev->data->dev_conf.intr_conf.rmv)
+		rte_eth_dev_callback_process(bp->eth_dev,
+					     RTE_ETH_EVENT_INTR_RMV,
+					     NULL);
 	PMD_DRV_LOG(ERR, "Failed to recover from FW reset\n");
 }
 
@@ -4405,7 +4446,7 @@ static int bnxt_alloc_ctx_mem_blk(struct bnxt *bp,
 		if (!mz) {
 			mz = rte_memzone_reserve_aligned(mz_name,
 						rmem->nr_pages * 8,
-						SOCKET_ID_ANY,
+						bp->eth_dev->device->numa_node,
 						RTE_MEMZONE_2MB |
 						RTE_MEMZONE_SIZE_HINT_ONLY |
 						RTE_MEMZONE_IOVA_CONTIG,
@@ -4428,7 +4469,7 @@ static int bnxt_alloc_ctx_mem_blk(struct bnxt *bp,
 	if (!mz) {
 		mz = rte_memzone_reserve_aligned(mz_name,
 						 mem_size,
-						 SOCKET_ID_ANY,
+						 bp->eth_dev->device->numa_node,
 						 RTE_MEMZONE_1GB |
 						 RTE_MEMZONE_SIZE_HINT_ONLY |
 						 RTE_MEMZONE_IOVA_CONTIG,
@@ -5652,7 +5693,8 @@ bnxt_dev_init(struct rte_eth_dev *eth_dev, void *params __rte_unused)
 		goto error_free;
 
 	PMD_DRV_LOG(INFO,
-		    DRV_MODULE_NAME "found at mem %" PRIX64 ", node addr %pM\n",
+		    "Found %s device at mem %" PRIX64 ", node addr %pM\n",
+		    DRV_MODULE_NAME,
 		    pci_dev->mem_resource[0].phys_addr,
 		    pci_dev->mem_resource[0].addr);
 
@@ -6106,6 +6148,7 @@ static int bnxt_pci_remove(struct rte_pci_device *pci_dev)
 static struct rte_pci_driver bnxt_rte_pmd = {
 	.id_table = bnxt_pci_id_map,
 	.drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC |
+			RTE_PCI_DRV_INTR_RMV |
 			RTE_PCI_DRV_PROBE_AGAIN, /* Needed in case of VF-REPs
 						  * and OVS-DPDK
 						  */
diff --git a/drivers/net/bnxt/bnxt_flow.c b/drivers/net/bnxt/bnxt_flow.c
index 11034b6..646d8dc 100644
--- a/drivers/net/bnxt/bnxt_flow.c
+++ b/drivers/net/bnxt/bnxt_flow.c
@@ -919,36 +919,59 @@ bnxt_get_l2_filter(struct bnxt *bp, struct bnxt_filter_info *nf,
 	return l2_filter;
 }
 
-static int bnxt_vnic_prep(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+static void bnxt_vnic_cleanup(struct bnxt *bp, struct bnxt_vnic_info *vnic)
+{
+	if (vnic->rx_queue_cnt > 1)
+		bnxt_hwrm_vnic_ctx_free(bp, vnic);
+
+	bnxt_hwrm_vnic_free(bp, vnic);
+
+	rte_free(vnic->fw_grp_ids);
+	vnic->fw_grp_ids = NULL;
+
+	vnic->rx_queue_cnt = 0;
+}
+
+static int bnxt_vnic_prep(struct bnxt *bp, struct bnxt_vnic_info *vnic,
+			  const struct rte_flow_action *act,
+			  struct rte_flow_error *error)
 {
 	struct rte_eth_conf *dev_conf = &bp->eth_dev->data->dev_conf;
 	uint64_t rx_offloads = dev_conf->rxmode.offloads;
 	int rc;
 
 	if (bp->nr_vnics > bp->max_vnics - 1)
-		return -ENOMEM;
+		return rte_flow_error_set(error, EINVAL,
+					  RTE_FLOW_ERROR_TYPE_ATTR_GROUP,
+					  NULL,
+					  "Group id is invalid");
 
 	rc = bnxt_vnic_grp_alloc(bp, vnic);
 	if (rc)
-		goto ret;
+		return rte_flow_error_set(error, -rc,
+					  RTE_FLOW_ERROR_TYPE_ACTION,
+					  act,
+					  "Failed to alloc VNIC group");
 
 	rc = bnxt_hwrm_vnic_alloc(bp, vnic);
 	if (rc) {
-		PMD_DRV_LOG(ERR, "HWRM vnic alloc failure rc: %x\n", rc);
+		rte_flow_error_set(error, -rc,
+				   RTE_FLOW_ERROR_TYPE_ACTION,
+				   act,
+				   "Failed to alloc VNIC");
 		goto ret;
 	}
-	bp->nr_vnics++;
 
 	/* RSS context is required only when there is more than one RSS ring */
 	if (vnic->rx_queue_cnt > 1) {
-		rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic, 0 /* ctx_idx 0 */);
+		rc = bnxt_hwrm_vnic_ctx_alloc(bp, vnic, 0);
 		if (rc) {
-			PMD_DRV_LOG(ERR,
-				    "HWRM vnic ctx alloc failure: %x\n", rc);
+			rte_flow_error_set(error, -rc,
+					   RTE_FLOW_ERROR_TYPE_ACTION,
+					   act,
+					   "Failed to alloc VNIC context");
 			goto ret;
 		}
-	} else {
-		PMD_DRV_LOG(DEBUG, "No RSS context required\n");
 	}
 
 	if (rx_offloads & DEV_RX_OFFLOAD_VLAN_STRIP)
@@ -957,12 +980,29 @@ static int bnxt_vnic_prep(struct bnxt *bp, struct bnxt_vnic_info *vnic)
 		vnic->vlan_strip = false;
 
 	rc = bnxt_hwrm_vnic_cfg(bp, vnic);
-	if (rc)
+	if (rc) {
+		rte_flow_error_set(error, -rc,
+				   RTE_FLOW_ERROR_TYPE_ACTION,
+				   act,
+				   "Failed to configure VNIC");
+		goto ret;
+	}
+
+	rc = bnxt_hwrm_vnic_plcmode_cfg(bp, vnic);
+	if (rc) {
+		rte_flow_error_set(error, -rc,
+				   RTE_FLOW_ERROR_TYPE_ACTION,
+				   act,
+				   "Failed to configure VNIC plcmode");
 		goto ret;
+	}
 
-	bnxt_hwrm_vnic_plcmode_cfg(bp, vnic);
+	bp->nr_vnics++;
+
+	return 0;
 
 ret:
+	bnxt_vnic_cleanup(bp, vnic);
 	return rc;
 }
 
@@ -1135,16 +1175,9 @@ start:
 
 		PMD_DRV_LOG(DEBUG, "VNIC found\n");
 
-		rc = bnxt_vnic_prep(bp, vnic);
-		if (rc)  {
-			rte_flow_error_set(error,
-					   EINVAL,
-					   RTE_FLOW_ERROR_TYPE_ACTION,
-					   act,
-					   "VNIC prep fail");
-			rc = -rte_errno;
+		rc = bnxt_vnic_prep(bp, vnic, act, error);
+		if (rc)
 			goto ret;
-		}
 
 		PMD_DRV_LOG(DEBUG,
 			    "vnic[%d] = %p vnic->fw_grp_ids = %p\n",
@@ -1355,16 +1388,9 @@ use_vnic:
 		vnic->end_grp_id = rss->queue[rss->queue_num - 1];
 		vnic->func_default = 0;	//This is not a default VNIC.
 
-		rc = bnxt_vnic_prep(bp, vnic);
-		if (rc) {
-			rte_flow_error_set(error,
-					   EINVAL,
-					   RTE_FLOW_ERROR_TYPE_ACTION,
-					   act,
-					   "VNIC prep fail");
-			rc = -rte_errno;
+		rc = bnxt_vnic_prep(bp, vnic, act, error);
+		if (rc)
 			goto ret;
-		}
 
 		PMD_DRV_LOG(DEBUG,
 			    "vnic[%d] = %p vnic->fw_grp_ids = %p\n",
@@ -1529,9 +1555,11 @@ bnxt_flow_validate(struct rte_eth_dev *dev,
 
 	filter = bnxt_get_unused_filter(bp);
 	if (filter == NULL) {
-		PMD_DRV_LOG(ERR, "Not enough resources for a new flow.\n");
+		rte_flow_error_set(error, ENOSPC,
+				   RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+				   "Not enough resources for a new flow");
 		bnxt_release_flow_lock(bp);
-		return -ENOMEM;
+		return -ENOSPC;
 	}
 
 	ret = bnxt_validate_and_parse_flow(dev, pattern, actions, attr,
@@ -1542,10 +1570,7 @@ bnxt_flow_validate(struct rte_eth_dev *dev,
 	vnic = find_matching_vnic(bp, filter);
 	if (vnic) {
 		if (STAILQ_EMPTY(&vnic->filter)) {
-			rte_free(vnic->fw_grp_ids);
-			bnxt_hwrm_vnic_ctx_free(bp, vnic);
-			bnxt_hwrm_vnic_free(bp, vnic);
-			vnic->rx_queue_cnt = 0;
+			bnxt_vnic_cleanup(bp, vnic);
 			bp->nr_vnics--;
 			PMD_DRV_LOG(DEBUG, "Free VNIC\n");
 		}
@@ -1931,12 +1956,20 @@ static int bnxt_handle_tunnel_redirect_destroy(struct bnxt *bp,
 		/* Tunnel doesn't belong to this VF, so don't send HWRM
 		 * cmd, just delete the flow from driver
 		 */
-		if (bp->fw_fid != (tun_dst_fid + bp->first_vf_id))
+		if (bp->fw_fid != (tun_dst_fid + bp->first_vf_id)) {
 			PMD_DRV_LOG(ERR,
 				    "Tunnel does not belong to this VF, skip hwrm_tunnel_redirect_free\n");
-		else
+		} else {
 			ret = bnxt_hwrm_tunnel_redirect_free(bp,
 							filter->tunnel_type);
+			if (ret) {
+				rte_flow_error_set(error, -ret,
+						   RTE_FLOW_ERROR_TYPE_HANDLE,
+						   NULL,
+						   "Unable to free tunnel redirection");
+				return ret;
+			}
+		}
 	}
 	return ret;
 }
@@ -1999,12 +2032,7 @@ done:
 		 */
 		if (vnic && !vnic->func_default &&
 		    STAILQ_EMPTY(&vnic->flow_list)) {
-			rte_free(vnic->fw_grp_ids);
-			if (vnic->rx_queue_cnt > 1)
-				bnxt_hwrm_vnic_ctx_free(bp, vnic);
-
-			bnxt_hwrm_vnic_free(bp, vnic);
-			vnic->rx_queue_cnt = 0;
+			bnxt_vnic_cleanup(bp, vnic);
 			bp->nr_vnics--;
 		}
 	} else {
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 5ed38c9..39778e3 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -635,9 +635,13 @@ static int bnxt_hwrm_ptp_qcfg(struct bnxt *bp)
 
 	HWRM_CHECK_RESULT();
 
-	if (!BNXT_CHIP_THOR(bp) &&
-	    !(resp->flags & HWRM_PORT_MAC_PTP_QCFG_OUTPUT_FLAGS_DIRECT_ACCESS))
-		return 0;
+	if (BNXT_CHIP_THOR(bp)) {
+		if (!(resp->flags & HWRM_PORT_MAC_PTP_QCFG_OUTPUT_FLAGS_HWRM_ACCESS))
+			return 0;
+	} else {
+		if (!(resp->flags & HWRM_PORT_MAC_PTP_QCFG_OUTPUT_FLAGS_DIRECT_ACCESS))
+			return 0;
+	}
 
 	if (resp->flags & HWRM_PORT_MAC_PTP_QCFG_OUTPUT_FLAGS_ONE_STEP_TX_TS)
 		bp->flags |= BNXT_FLAG_FW_CAP_ONE_STEP_TX_TS;
@@ -1359,6 +1363,7 @@ static int bnxt_hwrm_port_phy_qcfg(struct bnxt *bp,
 
 	link_info->support_speeds = rte_le_to_cpu_16(resp->support_speeds);
 	link_info->auto_link_speed = rte_le_to_cpu_16(resp->auto_link_speed);
+	link_info->auto_link_speed_mask = rte_le_to_cpu_16(resp->auto_link_speed_mask);
 	link_info->preemphasis = rte_le_to_cpu_32(resp->preemphasis);
 	link_info->force_link_speed = rte_le_to_cpu_16(resp->force_link_speed);
 	link_info->phy_ver[0] = resp->phy_maj;
@@ -1412,6 +1417,12 @@ int bnxt_hwrm_port_phy_qcaps(struct bnxt *bp)
 
 	HWRM_UNLOCK();
 
+	/* Older firmware does not have supported_auto_speeds, so assume
+	 * that all supported speeds can be autonegotiated.
+	 */
+	if (link_info->auto_link_speed_mask && !link_info->support_auto_speeds)
+		link_info->support_auto_speeds = link_info->support_speeds;
+
 	return 0;
 }
 
@@ -1665,12 +1676,16 @@ int bnxt_hwrm_ring_free(struct bnxt *bp,
 	struct hwrm_ring_free_input req = {.req_type = 0 };
 	struct hwrm_ring_free_output *resp = bp->hwrm_cmd_resp_addr;
 
+	if (ring->fw_ring_id == INVALID_HW_RING_ID)
+		return -EINVAL;
+
 	HWRM_PREP(&req, HWRM_RING_FREE, BNXT_USE_CHIMP_MB);
 
 	req.ring_type = ring_type;
 	req.ring_id = rte_cpu_to_le_16(ring->fw_ring_id);
 
 	rc = bnxt_hwrm_send_message(bp, &req, sizeof(req), BNXT_USE_CHIMP_MB);
+	ring->fw_ring_id = INVALID_HW_RING_ID;
 
 	if (rc || resp->error_code) {
 		if (rc == 0 && resp->error_code)
@@ -1756,7 +1771,7 @@ int bnxt_hwrm_stat_clear(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
 	struct hwrm_stat_ctx_clr_stats_input req = {.req_type = 0 };
 	struct hwrm_stat_ctx_clr_stats_output *resp = bp->hwrm_cmd_resp_addr;
 
-	if (cpr->hw_stats_ctx_id == (uint32_t)HWRM_NA_SIGNATURE)
+	if (cpr->hw_stats_ctx_id == HWRM_NA_SIGNATURE)
 		return rc;
 
 	HWRM_PREP(&req, HWRM_STAT_CTX_CLR_STATS, BNXT_USE_CHIMP_MB);
@@ -1777,6 +1792,9 @@ static int bnxt_hwrm_stat_ctx_alloc(struct bnxt *bp, struct bnxt_cp_ring_info *c
 	struct hwrm_stat_ctx_alloc_input req = {.req_type = 0 };
 	struct hwrm_stat_ctx_alloc_output *resp = bp->hwrm_cmd_resp_addr;
 
+	if (cpr->hw_stats_ctx_id != HWRM_NA_SIGNATURE)
+		return 0;
+
 	HWRM_PREP(&req, HWRM_STAT_CTX_ALLOC, BNXT_USE_CHIMP_MB);
 
 	req.update_period_ms = rte_cpu_to_le_32(0);
@@ -1800,6 +1818,9 @@ static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp, struct bnxt_cp_ring_info *cp
 	struct hwrm_stat_ctx_free_input req = {.req_type = 0 };
 	struct hwrm_stat_ctx_free_output *resp = bp->hwrm_cmd_resp_addr;
 
+	if (cpr->hw_stats_ctx_id == HWRM_NA_SIGNATURE)
+		return 0;
+
 	HWRM_PREP(&req, HWRM_STAT_CTX_FREE, BNXT_USE_CHIMP_MB);
 
 	req.stat_ctx_id = rte_cpu_to_le_32(cpr->hw_stats_ctx_id);
@@ -1809,6 +1830,8 @@ static int bnxt_hwrm_stat_ctx_free(struct bnxt *bp, struct bnxt_cp_ring_info *cp
 	HWRM_CHECK_RESULT();
 	HWRM_UNLOCK();
 
+	cpr->hw_stats_ctx_id = HWRM_NA_SIGNATURE;
+
 	return rc;
 }
 
@@ -2449,48 +2472,54 @@ bnxt_free_all_hwrm_stat_ctxs(struct bnxt *bp)
 	unsigned int i;
 	struct bnxt_cp_ring_info *cpr;
 
-	for (i = 0; i < bp->rx_cp_nr_rings + bp->tx_cp_nr_rings; i++) {
+	for (i = 0; i < bp->rx_cp_nr_rings; i++) {
 
-		if (i >= bp->rx_cp_nr_rings) {
-			cpr = bp->tx_queues[i - bp->rx_cp_nr_rings]->cp_ring;
-		} else {
-			cpr = bp->rx_queues[i]->cp_ring;
-			if (BNXT_HAS_RING_GRPS(bp))
-				bp->grp_info[i].fw_stats_ctx = -1;
-		}
-		if (cpr->hw_stats_ctx_id != HWRM_NA_SIGNATURE) {
-			rc = bnxt_hwrm_stat_ctx_free(bp, cpr);
-			cpr->hw_stats_ctx_id = HWRM_NA_SIGNATURE;
-			if (rc)
-				return rc;
-		}
+		cpr = bp->rx_queues[i]->cp_ring;
+		if (BNXT_HAS_RING_GRPS(bp))
+			bp->grp_info[i].fw_stats_ctx = -1;
+		rc = bnxt_hwrm_stat_ctx_free(bp, cpr);
+		if (rc)
+			return rc;
 	}
+
+	for (i = 0; i < bp->tx_cp_nr_rings; i++) {
+		cpr = bp->tx_queues[i]->cp_ring;
+		rc = bnxt_hwrm_stat_ctx_free(bp, cpr);
+		if (rc)
+			return rc;
+	}
+
 	return 0;
 }
 
 int bnxt_alloc_all_hwrm_stat_ctxs(struct bnxt *bp)
 {
+	struct bnxt_cp_ring_info *cpr;
 	unsigned int i;
 	int rc = 0;
 
-	for (i = 0; i < bp->rx_cp_nr_rings + bp->tx_cp_nr_rings; i++) {
-		struct bnxt_tx_queue *txq;
-		struct bnxt_rx_queue *rxq;
-		struct bnxt_cp_ring_info *cpr;
+	for (i = 0; i < bp->rx_cp_nr_rings; i++) {
+		struct bnxt_rx_queue *rxq = bp->rx_queues[i];
 
-		if (i >= bp->rx_cp_nr_rings) {
-			txq = bp->tx_queues[i - bp->rx_cp_nr_rings];
-			cpr = txq->cp_ring;
-		} else {
-			rxq = bp->rx_queues[i];
-			cpr = rxq->cp_ring;
+		cpr = rxq->cp_ring;
+		if (cpr->hw_stats_ctx_id == HWRM_NA_SIGNATURE) {
+			rc = bnxt_hwrm_stat_ctx_alloc(bp, cpr);
+			if (rc)
+				return rc;
 		}
+	}
 
-		rc = bnxt_hwrm_stat_ctx_alloc(bp, cpr);
+	for (i = 0; i < bp->tx_cp_nr_rings; i++) {
+		struct bnxt_tx_queue *txq = bp->tx_queues[i];
 
-		if (rc)
-			return rc;
+		cpr = txq->cp_ring;
+		if (cpr->hw_stats_ctx_id == HWRM_NA_SIGNATURE) {
+			rc = bnxt_hwrm_stat_ctx_alloc(bp, cpr);
+			if (rc)
+				return rc;
+		}
 	}
+
 	return rc;
 }
 
@@ -2523,10 +2552,9 @@ void bnxt_free_nq_ring(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
 	bnxt_hwrm_ring_free(bp, cp_ring,
 			    HWRM_RING_FREE_INPUT_RING_TYPE_NQ);
 	cp_ring->fw_ring_id = INVALID_HW_RING_ID;
-	memset(cpr->cp_desc_ring, 0, cpr->cp_ring_struct->ring_size *
-				     sizeof(*cpr->cp_desc_ring));
+	memset(cpr->cp_desc_ring, 0,
+	       cpr->cp_ring_struct->ring_size * sizeof(*cpr->cp_desc_ring));
 	cpr->cp_raw_cons = 0;
-	cpr->valid = 0;
 }
 
 void bnxt_free_cp_ring(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
@@ -2534,12 +2562,11 @@ void bnxt_free_cp_ring(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
 	struct bnxt_ring *cp_ring = cpr->cp_ring_struct;
 
 	bnxt_hwrm_ring_free(bp, cp_ring,
-			HWRM_RING_FREE_INPUT_RING_TYPE_L2_CMPL);
+			    HWRM_RING_FREE_INPUT_RING_TYPE_L2_CMPL);
 	cp_ring->fw_ring_id = INVALID_HW_RING_ID;
-	memset(cpr->cp_desc_ring, 0, cpr->cp_ring_struct->ring_size *
-			sizeof(*cpr->cp_desc_ring));
+	memset(cpr->cp_desc_ring, 0,
+	       cpr->cp_ring_struct->ring_size * sizeof(*cpr->cp_desc_ring));
 	cpr->cp_raw_cons = 0;
-	cpr->valid = 0;
 }
 
 void bnxt_free_hwrm_rx_ring(struct bnxt *bp, int queue_index)
@@ -2549,31 +2576,46 @@ void bnxt_free_hwrm_rx_ring(struct bnxt *bp, int queue_index)
 	struct bnxt_ring *ring = rxr->rx_ring_struct;
 	struct bnxt_cp_ring_info *cpr = rxq->cp_ring;
 
-	if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-		bnxt_hwrm_ring_free(bp, ring,
-				    HWRM_RING_FREE_INPUT_RING_TYPE_RX);
-		ring->fw_ring_id = INVALID_HW_RING_ID;
-		if (BNXT_HAS_RING_GRPS(bp))
-			bp->grp_info[queue_index].rx_fw_ring_id =
-							INVALID_HW_RING_ID;
-	}
+	bnxt_hwrm_ring_free(bp, ring,
+			    HWRM_RING_FREE_INPUT_RING_TYPE_RX);
+	if (BNXT_HAS_RING_GRPS(bp))
+		bp->grp_info[queue_index].rx_fw_ring_id =
+						INVALID_HW_RING_ID;
+
 	ring = rxr->ag_ring_struct;
-	if (ring->fw_ring_id != INVALID_HW_RING_ID) {
-		bnxt_hwrm_ring_free(bp, ring,
-				    BNXT_CHIP_THOR(bp) ?
-				    HWRM_RING_FREE_INPUT_RING_TYPE_RX_AGG :
-				    HWRM_RING_FREE_INPUT_RING_TYPE_RX);
-		if (BNXT_HAS_RING_GRPS(bp))
-			bp->grp_info[queue_index].ag_fw_ring_id =
-							INVALID_HW_RING_ID;
-	}
-	if (cpr->cp_ring_struct->fw_ring_id != INVALID_HW_RING_ID)
-		bnxt_free_cp_ring(bp, cpr);
+	bnxt_hwrm_ring_free(bp, ring,
+			    BNXT_CHIP_THOR(bp) ?
+			    HWRM_RING_FREE_INPUT_RING_TYPE_RX_AGG :
+			    HWRM_RING_FREE_INPUT_RING_TYPE_RX);
+	if (BNXT_HAS_RING_GRPS(bp))
+		bp->grp_info[queue_index].ag_fw_ring_id =
+						INVALID_HW_RING_ID;
+
+	bnxt_free_cp_ring(bp, cpr);
 
 	if (BNXT_HAS_RING_GRPS(bp))
 		bp->grp_info[queue_index].cp_fw_ring_id = INVALID_HW_RING_ID;
 }
 
+int bnxt_hwrm_rx_ring_reset(struct bnxt *bp, int queue_index)
+{
+	int rc;
+	struct hwrm_ring_reset_input req = {.req_type = 0 };
+	struct hwrm_ring_reset_output *resp = bp->hwrm_cmd_resp_addr;
+
+	HWRM_PREP(&req, HWRM_RING_RESET, BNXT_USE_CHIMP_MB);
+
+	req.ring_type = HWRM_RING_RESET_INPUT_RING_TYPE_RX_RING_GRP;
+	req.ring_id = rte_cpu_to_le_16(bp->grp_info[queue_index].fw_grp_id);
+	rc = bnxt_hwrm_send_message(bp, &req, sizeof(req), BNXT_USE_CHIMP_MB);
+
+	HWRM_CHECK_RESULT();
+
+	HWRM_UNLOCK();
+
+	return rc;
+}
+
 static int
 bnxt_free_all_hwrm_rings(struct bnxt *bp)
 {
@@ -3081,15 +3123,8 @@ int bnxt_set_hwrm_link_config(struct bnxt *bp, bool link_up)
 	speed = bnxt_parse_eth_link_speed(dev_conf->link_speeds,
 					  bp->link_info->link_signal_mode);
 	link_req.phy_flags = HWRM_PORT_PHY_CFG_INPUT_FLAGS_RESET_PHY;
-	/* Autoneg can be done only when the FW allows.
-	 * When user configures fixed speed of 40G and later changes to
-	 * any other speed, auto_link_speed/force_link_speed is still set
-	 * to 40G until link comes up at new speed.
-	 */
-	if (autoneg == 1 &&
-	    !(!BNXT_CHIP_THOR(bp) &&
-	      (bp->link_info->auto_link_speed ||
-	       bp->link_info->force_link_speed))) {
+	/* Autoneg can be done only when the FW allows. */
+	if (autoneg == 1 && bp->link_info->support_auto_speeds) {
 		link_req.phy_flags |=
 				HWRM_PORT_PHY_CFG_INPUT_FLAGS_RESTART_AUTONEG;
 		link_req.auto_link_speed_mask =
@@ -3145,7 +3180,6 @@ error:
 	return rc;
 }
 
-/* JIRA 22088 */
 int bnxt_hwrm_func_qcfg(struct bnxt *bp, uint16_t *mtu)
 {
 	struct hwrm_func_qcfg_input req = {0};
@@ -3162,8 +3196,7 @@ int bnxt_hwrm_func_qcfg(struct bnxt *bp, uint16_t *mtu)
 
 	HWRM_CHECK_RESULT();
 
-	/* Hard Coded.. 0xfff VLAN ID mask */
-	bp->vlan = rte_le_to_cpu_16(resp->vlan) & 0xfff;
+	bp->vlan = rte_le_to_cpu_16(resp->vlan) & ETH_VLAN_ID_MAX;
 
 	svif_info = rte_le_to_cpu_16(resp->svif_info);
 	if (svif_info & HWRM_FUNC_QCFG_OUTPUT_SVIF_INFO_SVIF_VALID)
@@ -3233,16 +3266,6 @@ int bnxt_hwrm_parent_pf_qcfg(struct bnxt *bp)
 	bp->parent->fid = rte_le_to_cpu_16(resp->fid);
 	bp->parent->port_id = rte_le_to_cpu_16(resp->port_id);
 
-	/* FIXME: Temporary workaround - remove when firmware issue is fixed. */
-	if (bp->parent->vnic == 0) {
-		PMD_DRV_LOG(DEBUG, "parent VNIC unavailable.\n");
-		/* Use hard-coded values appropriate for current Wh+ fw. */
-		if (bp->parent->fid == 2)
-			bp->parent->vnic = 0x100;
-		else
-			bp->parent->vnic = 1;
-	}
-
 	HWRM_UNLOCK();
 
 	return 0;
@@ -4145,8 +4168,20 @@ int bnxt_hwrm_exec_fwd_resp(struct bnxt *bp, uint16_t target_id,
 	return rc;
 }
 
-int bnxt_hwrm_ctx_qstats(struct bnxt *bp, uint32_t cid, int idx,
-			 struct rte_eth_stats *stats, uint8_t rx)
+static void bnxt_update_prev_stat(uint64_t *cntr, uint64_t *prev_cntr)
+{
+	/* One of the HW stat values that make up this counter was zero as
+	 * returned by HW in this iteration, so use the previous
+	 * iteration's counter value
+	 */
+	if (*prev_cntr && *cntr == 0)
+		*cntr = *prev_cntr;
+	else
+		*prev_cntr = *cntr;
+}
+
+int bnxt_hwrm_ring_stats(struct bnxt *bp, uint32_t cid, int idx,
+			 struct bnxt_ring_stats *ring_stats, bool rx)
 {
 	int rc = 0;
 	struct hwrm_stat_ctx_query_input req = {.req_type = 0};
@@ -4161,21 +4196,85 @@ int bnxt_hwrm_ctx_qstats(struct bnxt *bp, uint32_t cid, int idx,
 	HWRM_CHECK_RESULT();
 
 	if (rx) {
-		stats->q_ipackets[idx] = rte_le_to_cpu_64(resp->rx_ucast_pkts);
-		stats->q_ipackets[idx] += rte_le_to_cpu_64(resp->rx_mcast_pkts);
-		stats->q_ipackets[idx] += rte_le_to_cpu_64(resp->rx_bcast_pkts);
-		stats->q_ibytes[idx] = rte_le_to_cpu_64(resp->rx_ucast_bytes);
-		stats->q_ibytes[idx] += rte_le_to_cpu_64(resp->rx_mcast_bytes);
-		stats->q_ibytes[idx] += rte_le_to_cpu_64(resp->rx_bcast_bytes);
-		stats->q_errors[idx] = rte_le_to_cpu_64(resp->rx_discard_pkts);
-		stats->q_errors[idx] += rte_le_to_cpu_64(resp->rx_error_pkts);
+		struct bnxt_ring_stats *prev_stats = &bp->prev_rx_ring_stats[idx];
+
+		ring_stats->rx_ucast_pkts = rte_le_to_cpu_64(resp->rx_ucast_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_ucast_pkts,
+				      &prev_stats->rx_ucast_pkts);
+
+		ring_stats->rx_mcast_pkts = rte_le_to_cpu_64(resp->rx_mcast_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_mcast_pkts,
+				      &prev_stats->rx_mcast_pkts);
+
+		ring_stats->rx_bcast_pkts = rte_le_to_cpu_64(resp->rx_bcast_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_bcast_pkts,
+				      &prev_stats->rx_bcast_pkts);
+
+		ring_stats->rx_ucast_bytes = rte_le_to_cpu_64(resp->rx_ucast_bytes);
+		bnxt_update_prev_stat(&ring_stats->rx_ucast_bytes,
+				      &prev_stats->rx_ucast_bytes);
+
+		ring_stats->rx_mcast_bytes = rte_le_to_cpu_64(resp->rx_mcast_bytes);
+		bnxt_update_prev_stat(&ring_stats->rx_mcast_bytes,
+				      &prev_stats->rx_mcast_bytes);
+
+		ring_stats->rx_bcast_bytes = rte_le_to_cpu_64(resp->rx_bcast_bytes);
+		bnxt_update_prev_stat(&ring_stats->rx_bcast_bytes,
+				      &prev_stats->rx_bcast_bytes);
+
+		ring_stats->rx_discard_pkts = rte_le_to_cpu_64(resp->rx_discard_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_discard_pkts,
+				      &prev_stats->rx_discard_pkts);
+
+		ring_stats->rx_error_pkts = rte_le_to_cpu_64(resp->rx_error_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_error_pkts,
+				      &prev_stats->rx_error_pkts);
+
+		ring_stats->rx_agg_pkts = rte_le_to_cpu_64(resp->rx_agg_pkts);
+		bnxt_update_prev_stat(&ring_stats->rx_agg_pkts,
+				      &prev_stats->rx_agg_pkts);
+
+		ring_stats->rx_agg_bytes = rte_le_to_cpu_64(resp->rx_agg_bytes);
+		bnxt_update_prev_stat(&ring_stats->rx_agg_bytes,
+				      &prev_stats->rx_agg_bytes);
+
+		ring_stats->rx_agg_events = rte_le_to_cpu_64(resp->rx_agg_events);
+		bnxt_update_prev_stat(&ring_stats->rx_agg_events,
+				      &prev_stats->rx_agg_events);
+
+		ring_stats->rx_agg_aborts = rte_le_to_cpu_64(resp->rx_agg_aborts);
+		bnxt_update_prev_stat(&ring_stats->rx_agg_aborts,
+				      &prev_stats->rx_agg_aborts);
 	} else {
-		stats->q_opackets[idx] = rte_le_to_cpu_64(resp->tx_ucast_pkts);
-		stats->q_opackets[idx] += rte_le_to_cpu_64(resp->tx_mcast_pkts);
-		stats->q_opackets[idx] += rte_le_to_cpu_64(resp->tx_bcast_pkts);
-		stats->q_obytes[idx] = rte_le_to_cpu_64(resp->tx_ucast_bytes);
-		stats->q_obytes[idx] += rte_le_to_cpu_64(resp->tx_mcast_bytes);
-		stats->q_obytes[idx] += rte_le_to_cpu_64(resp->tx_bcast_bytes);
+		struct bnxt_ring_stats *prev_stats = &bp->prev_tx_ring_stats[idx];
+
+		ring_stats->tx_ucast_pkts = rte_le_to_cpu_64(resp->tx_ucast_pkts);
+		bnxt_update_prev_stat(&ring_stats->tx_ucast_pkts,
+				      &prev_stats->tx_ucast_pkts);
+
+		ring_stats->tx_mcast_pkts = rte_le_to_cpu_64(resp->tx_mcast_pkts);
+		bnxt_update_prev_stat(&ring_stats->tx_mcast_pkts,
+				      &prev_stats->tx_mcast_pkts);
+
+		ring_stats->tx_bcast_pkts = rte_le_to_cpu_64(resp->tx_bcast_pkts);
+		bnxt_update_prev_stat(&ring_stats->tx_bcast_pkts,
+				      &prev_stats->tx_bcast_pkts);
+
+		ring_stats->tx_ucast_bytes = rte_le_to_cpu_64(resp->tx_ucast_bytes);
+		bnxt_update_prev_stat(&ring_stats->tx_ucast_bytes,
+				      &prev_stats->tx_ucast_bytes);
+
+		ring_stats->tx_mcast_bytes = rte_le_to_cpu_64(resp->tx_mcast_bytes);
+		bnxt_update_prev_stat(&ring_stats->tx_mcast_bytes,
+				      &prev_stats->tx_mcast_bytes);
+
+		ring_stats->tx_bcast_bytes = rte_le_to_cpu_64(resp->tx_bcast_bytes);
+		bnxt_update_prev_stat(&ring_stats->tx_bcast_bytes,
+				      &prev_stats->tx_bcast_bytes);
+
+		ring_stats->tx_discard_pkts = rte_le_to_cpu_64(resp->tx_discard_pkts);
+		bnxt_update_prev_stat(&ring_stats->tx_discard_pkts,
+				      &prev_stats->tx_discard_pkts);
 	}
 
 	HWRM_UNLOCK();
@@ -5841,4 +5940,4 @@ int bnxt_hwrm_poll_ver_get(struct bnxt *bp)
 	HWRM_UNLOCK();
 
 	return rc;
-}
\ No newline at end of file
+}
diff --git a/drivers/net/bnxt/bnxt_hwrm.h b/drivers/net/bnxt/bnxt_hwrm.h
index a67a17b..82010e6 100644
--- a/drivers/net/bnxt/bnxt_hwrm.h
+++ b/drivers/net/bnxt/bnxt_hwrm.h
@@ -165,9 +165,6 @@ int bnxt_hwrm_ring_grp_alloc(struct bnxt *bp, unsigned int idx);
 int bnxt_hwrm_ring_grp_free(struct bnxt *bp, unsigned int idx);
 
 int bnxt_hwrm_stat_clear(struct bnxt *bp, struct bnxt_cp_ring_info *cpr);
-int bnxt_hwrm_ctx_qstats(struct bnxt *bp, uint32_t cid, int idx,
-			 struct rte_eth_stats *stats, uint8_t rx);
-
 int bnxt_hwrm_ver_get(struct bnxt *bp, uint32_t timeout);
 
 int bnxt_hwrm_vnic_alloc(struct bnxt *bp, struct bnxt_vnic_info *vnic);
@@ -297,4 +294,7 @@ int bnxt_hwrm_first_vf_id_query(struct bnxt *bp, uint16_t fid,
 int bnxt_hwrm_cfa_pair_alloc(struct bnxt *bp, struct bnxt_representor *rep);
 int bnxt_hwrm_cfa_pair_free(struct bnxt *bp, struct bnxt_representor *rep);
 int bnxt_hwrm_poll_ver_get(struct bnxt *bp);
+int bnxt_hwrm_ring_stats(struct bnxt *bp, uint32_t cid, int idx,
+			 struct bnxt_ring_stats *stats, bool rx);
+int bnxt_hwrm_rx_ring_reset(struct bnxt *bp, int queue_index);
 #endif
diff --git a/drivers/net/bnxt/bnxt_irq.c b/drivers/net/bnxt/bnxt_irq.c
index 40e1b0c..1a99508 100644
--- a/drivers/net/bnxt/bnxt_irq.c
+++ b/drivers/net/bnxt/bnxt_irq.c
@@ -21,11 +21,14 @@ void bnxt_int_handler(void *param)
 {
 	struct rte_eth_dev *eth_dev = (struct rte_eth_dev *)param;
 	struct bnxt *bp = eth_dev->data->dev_private;
-	struct bnxt_cp_ring_info *cpr = bp->async_cp_ring;
+	uint32_t cons, raw_cons, cp_ring_size;
+	struct bnxt_cp_ring_info *cpr;
 	struct cmpl_base *cmp;
-	uint32_t raw_cons;
-	uint32_t cons;
 
+
+	if (bp == NULL)
+		return;
+	cpr = bp->async_cp_ring;
 	if (cpr == NULL)
 		return;
 
@@ -42,10 +45,11 @@ void bnxt_int_handler(void *param)
 			return;
 		}
 
+		cp_ring_size = cpr->cp_ring_struct->ring_size;
 		cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
 		cmp = &cpr->cp_desc_ring[cons];
 
-		if (!CMP_VALID(cmp, raw_cons, cpr->cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(cmp, raw_cons, cp_ring_size))
 			break;
 
 		bnxt_event_hwrm_resp_handler(bp, cmp);
diff --git a/drivers/net/bnxt/bnxt_ring.c b/drivers/net/bnxt/bnxt_ring.c
index 94cf7d3..6834244 100644
--- a/drivers/net/bnxt/bnxt_ring.c
+++ b/drivers/net/bnxt/bnxt_ring.c
@@ -94,7 +94,7 @@ int bnxt_alloc_ring_grps(struct bnxt *bp)
  * tx bd ring - Only non-zero length if tx_ring_info is not NULL
  * rx bd ring - Only non-zero length if rx_ring_info is not NULL
  */
-int bnxt_alloc_rings(struct bnxt *bp, uint16_t qidx,
+int bnxt_alloc_rings(struct bnxt *bp, unsigned int socket_id, uint16_t qidx,
 			    struct bnxt_tx_queue *txq,
 			    struct bnxt_rx_queue *rxq,
 			    struct bnxt_cp_ring_info *cp_ring_info,
@@ -203,7 +203,7 @@ int bnxt_alloc_rings(struct bnxt *bp, uint16_t qidx,
 	mz = rte_memzone_lookup(mz_name);
 	if (!mz) {
 		mz = rte_memzone_reserve_aligned(mz_name, total_alloc_len,
-				SOCKET_ID_ANY,
+				socket_id,
 				RTE_MEMZONE_2MB |
 				RTE_MEMZONE_SIZE_HINT_ONLY |
 				RTE_MEMZONE_IOVA_CONTIG,
@@ -422,24 +422,23 @@ int bnxt_alloc_rxtx_nq_ring(struct bnxt *bp)
 	struct bnxt_cp_ring_info *nqr;
 	struct bnxt_ring *ring;
 	int ring_index = BNXT_NUM_ASYNC_CPR(bp);
-	unsigned int socket_id;
 	uint8_t ring_type;
 	int rc = 0;
 
 	if (!BNXT_HAS_NQ(bp) || bp->rxtx_nq_ring)
 		return 0;
 
-	socket_id = rte_lcore_to_socket_id(rte_get_main_lcore());
-
 	nqr = rte_zmalloc_socket("nqr",
 				 sizeof(struct bnxt_cp_ring_info),
-				 RTE_CACHE_LINE_SIZE, socket_id);
+				 RTE_CACHE_LINE_SIZE,
+				 bp->eth_dev->device->numa_node);
 	if (nqr == NULL)
 		return -ENOMEM;
 
 	ring = rte_zmalloc_socket("bnxt_cp_ring_struct",
 				  sizeof(struct bnxt_ring),
-				  RTE_CACHE_LINE_SIZE, socket_id);
+				  RTE_CACHE_LINE_SIZE,
+				  bp->eth_dev->device->numa_node);
 	if (ring == NULL) {
 		rte_free(nqr);
 		return -ENOMEM;
@@ -454,7 +453,8 @@ int bnxt_alloc_rxtx_nq_ring(struct bnxt *bp)
 	ring->fw_ring_id = INVALID_HW_RING_ID;
 
 	nqr->cp_ring_struct = ring;
-	rc = bnxt_alloc_rings(bp, 0, NULL, NULL, nqr, NULL, "l2_nqr");
+	rc = bnxt_alloc_rings(bp, bp->eth_dev->device->numa_node, 0, NULL,
+			      NULL, nqr, NULL, "l2_nqr");
 	if (rc) {
 		rte_free(ring);
 		rte_free(nqr);
@@ -837,22 +837,21 @@ int bnxt_alloc_async_ring_struct(struct bnxt *bp)
 {
 	struct bnxt_cp_ring_info *cpr = NULL;
 	struct bnxt_ring *ring = NULL;
-	unsigned int socket_id;
 
 	if (BNXT_NUM_ASYNC_CPR(bp) == 0)
 		return 0;
 
-	socket_id = rte_lcore_to_socket_id(rte_get_main_lcore());
-
 	cpr = rte_zmalloc_socket("cpr",
 				 sizeof(struct bnxt_cp_ring_info),
-				 RTE_CACHE_LINE_SIZE, socket_id);
+				 RTE_CACHE_LINE_SIZE,
+				 bp->eth_dev->device->numa_node);
 	if (cpr == NULL)
 		return -ENOMEM;
 
 	ring = rte_zmalloc_socket("bnxt_cp_ring_struct",
 				  sizeof(struct bnxt_ring),
-				  RTE_CACHE_LINE_SIZE, socket_id);
+				  RTE_CACHE_LINE_SIZE,
+				  bp->eth_dev->device->numa_node);
 	if (ring == NULL) {
 		rte_free(cpr);
 		return -ENOMEM;
@@ -868,7 +867,6 @@ int bnxt_alloc_async_ring_struct(struct bnxt *bp)
 	bp->async_cp_ring = cpr;
 	cpr->cp_ring_struct = ring;
 
-	return bnxt_alloc_rings(bp, 0, NULL, NULL,
-				bp->async_cp_ring, NULL,
-				"def_cp");
+	return bnxt_alloc_rings(bp, bp->eth_dev->device->numa_node, 0, NULL,
+				NULL, bp->async_cp_ring, NULL, "def_cp");
 }
diff --git a/drivers/net/bnxt/bnxt_ring.h b/drivers/net/bnxt/bnxt_ring.h
index 0a4685d..201b391 100644
--- a/drivers/net/bnxt/bnxt_ring.h
+++ b/drivers/net/bnxt/bnxt_ring.h
@@ -66,7 +66,7 @@ struct bnxt_rx_ring_info;
 struct bnxt_cp_ring_info;
 void bnxt_free_ring(struct bnxt_ring *ring);
 int bnxt_alloc_ring_grps(struct bnxt *bp);
-int bnxt_alloc_rings(struct bnxt *bp, uint16_t qidx,
+int bnxt_alloc_rings(struct bnxt *bp, unsigned int socket_id, uint16_t qidx,
 			    struct bnxt_tx_queue *txq,
 			    struct bnxt_rx_queue *rxq,
 			    struct bnxt_cp_ring_info *cp_ring_info,
diff --git a/drivers/net/bnxt/bnxt_rxq.c b/drivers/net/bnxt/bnxt_rxq.c
index ffb7193..a2feb52 100644
--- a/drivers/net/bnxt/bnxt_rxq.c
+++ b/drivers/net/bnxt/bnxt_rxq.c
@@ -335,8 +335,8 @@ int bnxt_rx_queue_setup_op(struct rte_eth_dev *eth_dev,
 
 	eth_dev->data->rx_queues[queue_idx] = rxq;
 	/* Allocate RX ring hardware descriptors */
-	rc = bnxt_alloc_rings(bp, queue_idx, NULL, rxq, rxq->cp_ring, NULL,
-			     "rxr");
+	rc = bnxt_alloc_rings(bp, socket_id, queue_idx, NULL, rxq, rxq->cp_ring,
+			      NULL, "rxr");
 	if (rc) {
 		PMD_DRV_LOG(ERR,
 			    "ring_dma_zone_reserve for rx_ring failed!\n");
diff --git a/drivers/net/bnxt/bnxt_rxq.h b/drivers/net/bnxt/bnxt_rxq.h
index c72105c..980a16a 100644
--- a/drivers/net/bnxt/bnxt_rxq.h
+++ b/drivers/net/bnxt/bnxt_rxq.h
@@ -30,6 +30,7 @@ struct bnxt_rx_queue {
 	uint8_t			rx_deferred_start; /* not in global dev start */
 	uint8_t			rx_started; /* RX queue is started */
 	uint8_t			drop_en; /* Drop when rx desc not available. */
+	uint8_t			in_reset; /* Rx ring is scheduled for reset */
 
 	struct bnxt		*bp;
 	int			index;
diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
index d16340a..41be742 100644
--- a/drivers/net/bnxt/bnxt_rxr.c
+++ b/drivers/net/bnxt/bnxt_rxr.c
@@ -10,6 +10,7 @@
 #include <rte_byteorder.h>
 #include <rte_malloc.h>
 #include <rte_memory.h>
+#include <rte_alarm.h>
 
 #include "bnxt.h"
 #include "bnxt_reps.h"
@@ -17,9 +18,7 @@
 #include "bnxt_rxr.h"
 #include "bnxt_rxq.h"
 #include "hsi_struct_def_dpdk.h"
-#ifdef RTE_LIBRTE_IEEE1588
 #include "bnxt_hwrm.h"
-#endif
 
 #include <bnxt_tf_common.h>
 #include <ulp_mark_mgr.h>
@@ -127,6 +126,50 @@ struct rte_mbuf *bnxt_consume_rx_buf(struct bnxt_rx_ring_info *rxr,
 	return mbuf;
 }
 
+static void bnxt_rx_ring_reset(void *arg)
+{
+	struct bnxt *bp = arg;
+	int i, rc = 0;
+	struct bnxt_rx_queue *rxq;
+
+
+	for (i = 0; i < (int)bp->rx_nr_rings; i++) {
+		struct bnxt_rx_ring_info *rxr;
+
+		rxq = bp->rx_queues[i];
+		if (!rxq || !rxq->in_reset)
+			continue;
+
+		rxr = rxq->rx_ring;
+		/* Disable and flush TPA before resetting the RX ring */
+		if (rxr->tpa_info)
+			bnxt_hwrm_vnic_tpa_cfg(bp, rxq->vnic, false);
+		rc = bnxt_hwrm_rx_ring_reset(bp, i);
+		if (rc) {
+			PMD_DRV_LOG(ERR, "Rx ring%d reset failed\n", i);
+			continue;
+		}
+
+		bnxt_rx_queue_release_mbufs(rxq);
+		rxr->rx_prod = 0;
+		rxr->ag_prod = 0;
+		rxr->rx_next_cons = 0;
+		bnxt_init_one_rx_ring(rxq);
+		bnxt_db_write(&rxr->rx_db, rxr->rx_prod);
+		bnxt_db_write(&rxr->ag_db, rxr->ag_prod);
+		if (rxr->tpa_info)
+			bnxt_hwrm_vnic_tpa_cfg(bp, rxq->vnic, true);
+
+		rxq->in_reset = 0;
+	}
+}
+
+static void bnxt_sched_ring_reset(struct bnxt_rx_queue *rxq)
+{
+	rxq->in_reset = 1;
+	rte_eal_alarm_set(1, bnxt_rx_ring_reset, (void *)rxq->bp);
+}
+
 static void bnxt_tpa_start(struct bnxt_rx_queue *rxq,
 			   struct rx_tpa_start_cmpl *tpa_start,
 			   struct rx_tpa_start_cmpl_hi *tpa_start1)
@@ -141,6 +184,12 @@ static void bnxt_tpa_start(struct bnxt_rx_queue *rxq,
 
 	data_cons = tpa_start->opaque;
 	tpa_info = &rxr->tpa_info[agg_id];
+	if (unlikely(data_cons != rxr->rx_next_cons)) {
+		PMD_DRV_LOG(ERR, "TPA cons %x, expected cons %x\n",
+			    data_cons, rxr->rx_next_cons);
+		bnxt_sched_ring_reset(rxq);
+		return;
+	}
 
 	mbuf = bnxt_consume_rx_buf(rxr, data_cons);
 
@@ -177,6 +226,8 @@ static void bnxt_tpa_start(struct bnxt_rx_queue *rxq,
 	/* recycle next mbuf */
 	data_cons = RING_NEXT(rxr->rx_ring_struct, data_cons);
 	bnxt_reuse_rx_mbuf(rxr, bnxt_consume_rx_buf(rxr, data_cons));
+
+	rxr->rx_next_cons = RING_NEXT(rxr->rx_ring_struct, data_cons);
 }
 
 static int bnxt_agg_bufs_valid(struct bnxt_cp_ring_info *cpr,
@@ -191,7 +242,8 @@ static int bnxt_agg_bufs_valid(struct bnxt_cp_ring_info *cpr,
 	cpr->valid = FLIP_VALID(raw_cp_cons,
 				cpr->cp_ring_struct->ring_mask,
 				cpr->valid);
-	return CMP_VALID(agg_cmpl, raw_cp_cons, cpr->cp_ring_struct);
+	return bnxt_cpr_cmp_valid(agg_cmpl, raw_cp_cons,
+				  cpr->cp_ring_struct->ring_size);
 }
 
 /* TPA consume agg buffer out of order, allocate connected data only */
@@ -272,6 +324,34 @@ static int bnxt_rx_pages(struct bnxt_rx_queue *rxq,
 	return 0;
 }
 
+static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
+			   uint32_t *raw_cons, void *cmp)
+{
+	struct rx_pkt_cmpl *rxcmp = cmp;
+	uint32_t tmp_raw_cons = *raw_cons;
+	uint8_t cmp_type, agg_bufs = 0;
+
+	cmp_type = CMP_TYPE(rxcmp);
+
+	if (cmp_type == CMPL_BASE_TYPE_RX_L2) {
+		agg_bufs = BNXT_RX_L2_AGG_BUFS(rxcmp);
+	} else if (cmp_type == RX_TPA_END_CMPL_TYPE_RX_TPA_END) {
+		struct rx_tpa_end_cmpl *tpa_end = cmp;
+
+		if (BNXT_CHIP_THOR(bp))
+			return 0;
+
+		agg_bufs = BNXT_TPA_END_AGG_BUFS(tpa_end);
+	}
+
+	if (agg_bufs) {
+		if (!bnxt_agg_bufs_valid(cpr, agg_bufs, tmp_raw_cons))
+			return -EBUSY;
+	}
+	*raw_cons = tmp_raw_cons;
+	return 0;
+}
+
 static inline struct rte_mbuf *bnxt_tpa_end(
 		struct bnxt_rx_queue *rxq,
 		uint32_t *raw_cp_cons,
@@ -286,6 +366,13 @@ static inline struct rte_mbuf *bnxt_tpa_end(
 	uint8_t payload_offset;
 	struct bnxt_tpa_info *tpa_info;
 
+	if (unlikely(rxq->in_reset)) {
+		PMD_DRV_LOG(ERR, "rxq->in_reset: raw_cp_cons:%d\n",
+			    *raw_cp_cons);
+		bnxt_discard_rx(rxq->bp, cpr, raw_cp_cons, tpa_end);
+		return NULL;
+	}
+
 	if (BNXT_CHIP_THOR(rxq->bp)) {
 		struct rx_tpa_v2_end_cmpl *th_tpa_end;
 		struct rx_tpa_v2_end_cmpl_hi *th_tpa_end1;
@@ -741,7 +828,8 @@ static int bnxt_rx_pkt(struct rte_mbuf **rx_pkt,
 	cp_cons = RING_CMP(cpr->cp_ring_struct, tmp_raw_cons);
 	rxcmp1 = (struct rx_pkt_cmpl_hi *)&cpr->cp_desc_ring[cp_cons];
 
-	if (!CMP_VALID(rxcmp1, tmp_raw_cons, cpr->cp_ring_struct))
+	if (!bnxt_cpr_cmp_valid(rxcmp1, tmp_raw_cons,
+				cpr->cp_ring_struct->ring_size))
 		return -EBUSY;
 
 	cpr->valid = FLIP_VALID(cp_cons,
@@ -773,6 +861,14 @@ static int bnxt_rx_pkt(struct rte_mbuf **rx_pkt,
 	prod = rxr->rx_prod;
 
 	cons = rxcmp->opaque;
+	if (unlikely(cons != rxr->rx_next_cons)) {
+		bnxt_discard_rx(bp, cpr, &tmp_raw_cons, rxcmp);
+		PMD_DRV_LOG(ERR, "RX cons %x != expected cons %x\n",
+			    cons, rxr->rx_next_cons);
+		bnxt_sched_ring_reset(rxq);
+		rc = -EBUSY;
+		goto next_rx;
+	}
 	mbuf = bnxt_consume_rx_buf(rxr, cons);
 	if (mbuf == NULL)
 		return -EBUSY;
@@ -837,6 +933,7 @@ static int bnxt_rx_pkt(struct rte_mbuf **rx_pkt,
 		goto rx;
 	}
 	rxr->rx_prod = prod;
+	rxr->rx_next_cons = RING_NEXT(rxr->rx_ring_struct, cons);
 
 	if (BNXT_TRUFLOW_EN(bp) && (BNXT_VF_IS_TRUSTED(bp) || BNXT_PF(bp)) &&
 	    vfr_flag) {
@@ -911,7 +1008,8 @@ uint16_t bnxt_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
 		cons = RING_CMP(cpr->cp_ring_struct, raw_cons);
 		rxcmp = (struct rx_pkt_cmpl *)&cpr->cp_desc_ring[cons];
 
-		if (!CMP_VALID(rxcmp, raw_cons, cpr->cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(rxcmp, raw_cons,
+					cpr->cp_ring_struct->ring_size))
 			break;
 		cpr->valid = FLIP_VALID(cons,
 					cpr->cp_ring_struct->ring_mask,
diff --git a/drivers/net/bnxt/bnxt_rxr.h b/drivers/net/bnxt/bnxt_rxr.h
index f8eb9b7..46b2840 100644
--- a/drivers/net/bnxt/bnxt_rxr.h
+++ b/drivers/net/bnxt/bnxt_rxr.h
@@ -58,6 +58,7 @@ struct bnxt_rx_ring_info {
 	uint16_t		rx_prod;
 	uint16_t		ag_prod;
 	uint16_t                rx_cons; /* Needed for representor */
+	uint16_t                rx_next_cons;
 	struct bnxt_db_info     rx_db;
 	struct bnxt_db_info     ag_db;
 
diff --git a/drivers/net/bnxt/bnxt_rxtx_vec_neon.c b/drivers/net/bnxt/bnxt_rxtx_vec_neon.c
index 54f47a3..3cb9492 100644
--- a/drivers/net/bnxt/bnxt_rxtx_vec_neon.c
+++ b/drivers/net/bnxt/bnxt_rxtx_vec_neon.c
@@ -151,9 +151,8 @@ descs_to_mbufs(uint32x4_t mm_rxcmp[4], uint32x4_t mm_rxcmp1[4],
 	vst1q_u32((uint32_t *)&mbuf[3]->rx_descriptor_fields1, tmp);
 }
 
-uint16_t
-bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
-		   uint16_t nb_pkts)
+static uint16_t
+recv_burst_vec_neon(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
 {
 	struct bnxt_rx_queue *rxq = rx_queue;
 	struct bnxt_cp_ring_info *cpr = rxq->cp_ring;
@@ -178,9 +177,6 @@ bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	if (rxq->rxrearm_nb >= rxq->rx_free_thresh)
 		bnxt_rxq_rearm(rxq, rxr);
 
-	/* Return no more than RTE_BNXT_MAX_RX_BURST per call. */
-	nb_pkts = RTE_MIN(nb_pkts, RTE_BNXT_MAX_RX_BURST);
-
 	cons = raw_cons & (cp_ring_size - 1);
 	mbcons = (raw_cons / 2) & (rx_ring_size - 1);
 
@@ -314,6 +310,27 @@ out:
 	return nb_rx_pkts;
 }
 
+uint16_t
+bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+{
+	uint16_t cnt = 0;
+
+	while (nb_pkts > RTE_BNXT_MAX_RX_BURST) {
+		uint16_t burst;
+
+		burst = recv_burst_vec_neon(rx_queue, rx_pkts + cnt,
+					    RTE_BNXT_MAX_RX_BURST);
+
+		cnt += burst;
+		nb_pkts -= burst;
+
+		if (burst < RTE_BNXT_MAX_RX_BURST)
+			return cnt;
+	}
+
+	return cnt + recv_burst_vec_neon(rx_queue, rx_pkts + cnt, nb_pkts);
+}
+
 static void
 bnxt_handle_tx_cp_vec(struct bnxt_tx_queue *txq)
 {
@@ -330,7 +347,7 @@ bnxt_handle_tx_cp_vec(struct bnxt_tx_queue *txq)
 		cons = RING_CMPL(ring_mask, raw_cons);
 		txcmp = (struct tx_cmpl *)&cp_desc_ring[cons];
 
-		if (!CMP_VALID(txcmp, raw_cons, cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(txcmp, raw_cons, ring_mask + 1))
 			break;
 
 		if (likely(CMP_TYPE(txcmp) == TX_CMPL_TYPE_TX_L2))
diff --git a/drivers/net/bnxt/bnxt_rxtx_vec_sse.c b/drivers/net/bnxt/bnxt_rxtx_vec_sse.c
index 621f567..5974b47 100644
--- a/drivers/net/bnxt/bnxt_rxtx_vec_sse.c
+++ b/drivers/net/bnxt/bnxt_rxtx_vec_sse.c
@@ -143,9 +143,8 @@ descs_to_mbufs(__m128i mm_rxcmp[4], __m128i mm_rxcmp1[4],
 	_mm_store_si128((void *)&mbuf[3]->rx_descriptor_fields1, t0);
 }
 
-uint16_t
-bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
-		   uint16_t nb_pkts)
+static uint16_t
+recv_burst_vec_sse(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
 {
 	struct bnxt_rx_queue *rxq = rx_queue;
 	const __m128i mbuf_init = _mm_set_epi64x(0, rxq->mbuf_initializer);
@@ -170,9 +169,6 @@ bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts,
 	if (rxq->rxrearm_nb >= rxq->rx_free_thresh)
 		bnxt_rxq_rearm(rxq, rxr);
 
-	/* Return no more than RTE_BNXT_MAX_RX_BURST per call. */
-	nb_pkts = RTE_MIN(nb_pkts, RTE_BNXT_MAX_RX_BURST);
-
 	cons = raw_cons & (cp_ring_size - 1);
 	mbcons = (raw_cons / 2) & (rx_ring_size - 1);
 
@@ -296,6 +292,27 @@ out:
 	return nb_rx_pkts;
 }
 
+uint16_t
+bnxt_recv_pkts_vec(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+{
+	uint16_t cnt = 0;
+
+	while (nb_pkts > RTE_BNXT_MAX_RX_BURST) {
+		uint16_t burst;
+
+		burst = recv_burst_vec_sse(rx_queue, rx_pkts + cnt,
+					   RTE_BNXT_MAX_RX_BURST);
+
+		cnt += burst;
+		nb_pkts -= burst;
+
+		if (burst < RTE_BNXT_MAX_RX_BURST)
+			return cnt;
+	}
+
+	return cnt + recv_burst_vec_sse(rx_queue, rx_pkts + cnt, nb_pkts);
+}
+
 static void
 bnxt_handle_tx_cp_vec(struct bnxt_tx_queue *txq)
 {
@@ -312,7 +329,7 @@ bnxt_handle_tx_cp_vec(struct bnxt_tx_queue *txq)
 		cons = RING_CMPL(ring_mask, raw_cons);
 		txcmp = (struct tx_cmpl *)&cp_desc_ring[cons];
 
-		if (!CMP_VALID(txcmp, raw_cons, cp_ring_struct))
+		if (!bnxt_cpr_cmp_valid(txcmp, raw_cons, ring_mask + 1))
 			break;
 
 		if (likely(CMP_TYPE(txcmp) == TX_CMPL_TYPE_TX_L2))
diff --git a/drivers/net/bnxt/bnxt_stats.c b/drivers/net/bnxt/bnxt_stats.c
index 0cf3ee7..47dcf40 100644
--- a/drivers/net/bnxt/bnxt_stats.c
+++ b/drivers/net/bnxt/bnxt_stats.c
@@ -506,8 +506,47 @@ void bnxt_free_stats(struct bnxt *bp)
 	}
 }
 
+static void bnxt_fill_rte_eth_stats(struct rte_eth_stats *stats,
+				    struct bnxt_ring_stats *ring_stats,
+				    unsigned int i, bool rx)
+{
+	if (rx) {
+		stats->q_ipackets[i] = ring_stats->rx_ucast_pkts;
+		stats->q_ipackets[i] += ring_stats->rx_mcast_pkts;
+		stats->q_ipackets[i] += ring_stats->rx_bcast_pkts;
+
+		stats->ipackets += stats->q_ipackets[i];
+
+		stats->q_ibytes[i] = ring_stats->rx_ucast_bytes;
+		stats->q_ibytes[i] += ring_stats->rx_mcast_bytes;
+		stats->q_ibytes[i] += ring_stats->rx_bcast_bytes;
+
+		stats->ibytes += stats->q_ibytes[i];
+
+		stats->q_errors[i] = ring_stats->rx_discard_pkts;
+		stats->q_errors[i] += ring_stats->rx_error_pkts;
+
+		stats->imissed += ring_stats->rx_discard_pkts;
+		stats->ierrors += ring_stats->rx_error_pkts;
+	} else {
+		stats->q_opackets[i] = ring_stats->tx_ucast_pkts;
+		stats->q_opackets[i] += ring_stats->tx_mcast_pkts;
+		stats->q_opackets[i] += ring_stats->tx_bcast_pkts;
+
+		stats->opackets += stats->q_opackets[i];
+
+		stats->q_obytes[i] = ring_stats->tx_ucast_bytes;
+		stats->q_obytes[i] += ring_stats->tx_mcast_bytes;
+		stats->q_obytes[i] += ring_stats->tx_bcast_bytes;
+
+		stats->obytes += stats->q_obytes[i];
+
+		stats->oerrors += ring_stats->tx_discard_pkts;
+	}
+}
+
 int bnxt_stats_get_op(struct rte_eth_dev *eth_dev,
-			   struct rte_eth_stats *bnxt_stats)
+		      struct rte_eth_stats *bnxt_stats)
 {
 	int rc = 0;
 	unsigned int i;
@@ -527,11 +566,14 @@ int bnxt_stats_get_op(struct rte_eth_dev *eth_dev,
 	for (i = 0; i < num_q_stats; i++) {
 		struct bnxt_rx_queue *rxq = bp->rx_queues[i];
 		struct bnxt_cp_ring_info *cpr = rxq->cp_ring;
+		struct bnxt_ring_stats ring_stats = {0};
 
-		rc = bnxt_hwrm_ctx_qstats(bp, cpr->hw_stats_ctx_id, i,
-				     bnxt_stats, 1);
+		rc = bnxt_hwrm_ring_stats(bp, cpr->hw_stats_ctx_id, i,
+					  &ring_stats, true);
 		if (unlikely(rc))
 			return rc;
+
+		bnxt_fill_rte_eth_stats(bnxt_stats, &ring_stats, i, true);
 		bnxt_stats->rx_nombuf +=
 				rte_atomic64_read(&rxq->rx_mbuf_alloc_fail);
 	}
@@ -542,17 +584,29 @@ int bnxt_stats_get_op(struct rte_eth_dev *eth_dev,
 	for (i = 0; i < num_q_stats; i++) {
 		struct bnxt_tx_queue *txq = bp->tx_queues[i];
 		struct bnxt_cp_ring_info *cpr = txq->cp_ring;
+		struct bnxt_ring_stats ring_stats = {0};
 
-		rc = bnxt_hwrm_ctx_qstats(bp, cpr->hw_stats_ctx_id, i,
-				     bnxt_stats, 0);
+		rc = bnxt_hwrm_ring_stats(bp, cpr->hw_stats_ctx_id, i,
+					  &ring_stats, false);
 		if (unlikely(rc))
 			return rc;
+
+		bnxt_fill_rte_eth_stats(bnxt_stats, &ring_stats, i, false);
 	}
 
-	rc = bnxt_hwrm_func_qstats(bp, 0xffff, bnxt_stats, NULL);
 	return rc;
 }
 
+static void bnxt_clear_prev_stat(struct bnxt *bp)
+{
+	/*
+	 * Clear the cached values of stats returned by HW in the previous
+	 * get operation.
+	 */
+	memset(bp->prev_rx_ring_stats, 0, sizeof(struct bnxt_ring_stats) * bp->rx_cp_nr_rings);
+	memset(bp->prev_tx_ring_stats, 0, sizeof(struct bnxt_ring_stats) * bp->tx_cp_nr_rings);
+}
+
 int bnxt_stats_reset_op(struct rte_eth_dev *eth_dev)
 {
 	struct bnxt *bp = eth_dev->data->dev_private;
@@ -575,9 +629,45 @@ int bnxt_stats_reset_op(struct rte_eth_dev *eth_dev)
 		rte_atomic64_clear(&rxq->rx_mbuf_alloc_fail);
 	}
 
+	bnxt_clear_prev_stat(bp);
+
 	return ret;
 }
 
+static void bnxt_fill_func_qstats(struct hwrm_func_qstats_output *func_qstats,
+				  struct bnxt_ring_stats *ring_stats,
+				  bool rx)
+{
+	if (rx) {
+		func_qstats->rx_ucast_pkts += ring_stats->rx_ucast_pkts;
+		func_qstats->rx_mcast_pkts += ring_stats->rx_mcast_pkts;
+		func_qstats->rx_bcast_pkts += ring_stats->rx_bcast_pkts;
+
+		func_qstats->rx_ucast_bytes += ring_stats->rx_ucast_bytes;
+		func_qstats->rx_mcast_bytes += ring_stats->rx_mcast_bytes;
+		func_qstats->rx_bcast_bytes += ring_stats->rx_bcast_bytes;
+
+		func_qstats->rx_discard_pkts += ring_stats->rx_discard_pkts;
+		func_qstats->rx_drop_pkts += ring_stats->rx_error_pkts;
+
+		func_qstats->rx_agg_pkts += ring_stats->rx_agg_pkts;
+		func_qstats->rx_agg_bytes += ring_stats->rx_agg_bytes;
+		func_qstats->rx_agg_events += ring_stats->rx_agg_events;
+		func_qstats->rx_agg_aborts += ring_stats->rx_agg_aborts;
+	} else {
+		func_qstats->tx_ucast_pkts += ring_stats->tx_ucast_pkts;
+		func_qstats->tx_mcast_pkts += ring_stats->tx_mcast_pkts;
+		func_qstats->tx_bcast_pkts += ring_stats->tx_bcast_pkts;
+
+		func_qstats->tx_ucast_bytes += ring_stats->tx_ucast_bytes;
+		func_qstats->tx_mcast_bytes += ring_stats->tx_mcast_bytes;
+		func_qstats->tx_bcast_bytes += ring_stats->tx_bcast_bytes;
+
+		func_qstats->tx_drop_pkts += ring_stats->tx_error_pkts;
+		func_qstats->tx_discard_pkts += ring_stats->tx_discard_pkts;
+	}
+}
+
 int bnxt_dev_xstats_get_op(struct rte_eth_dev *eth_dev,
 			   struct rte_eth_xstat *xstats, unsigned int n)
 {
@@ -604,7 +694,38 @@ int bnxt_dev_xstats_get_op(struct rte_eth_dev *eth_dev,
 	if (n < stat_count || xstats == NULL)
 		return stat_count;
 
-	bnxt_hwrm_func_qstats(bp, 0xffff, NULL, &func_qstats);
+	for (i = 0; i < bp->rx_cp_nr_rings; i++) {
+		struct bnxt_rx_queue *rxq = bp->rx_queues[i];
+		struct bnxt_cp_ring_info *cpr = rxq->cp_ring;
+		struct bnxt_ring_stats ring_stats = {0};
+
+		if (!rxq->rx_started)
+			continue;
+
+		rc = bnxt_hwrm_ring_stats(bp, cpr->hw_stats_ctx_id, i,
+					  &ring_stats, true);
+		if (unlikely(rc))
+			return rc;
+
+		bnxt_fill_func_qstats(&func_qstats, &ring_stats, true);
+	}
+
+	for (i = 0; i < bp->tx_cp_nr_rings; i++) {
+		struct bnxt_tx_queue *txq = bp->tx_queues[i];
+		struct bnxt_cp_ring_info *cpr = txq->cp_ring;
+		struct bnxt_ring_stats ring_stats = {0};
+
+		if (!txq->tx_started)
+			continue;
+
+		rc = bnxt_hwrm_ring_stats(bp, cpr->hw_stats_ctx_id, i,
+					  &ring_stats, false);
+		if (unlikely(rc))
+			return rc;
+
+		bnxt_fill_func_qstats(&func_qstats, &ring_stats, false);
+	}
+
 	bnxt_hwrm_port_qstats(bp);
 	bnxt_hwrm_ext_port_qstats(bp);
 	rx_port_stats_ext_cnt = RTE_MIN(RTE_DIM(bnxt_rx_ext_stats_strings),
@@ -637,13 +758,11 @@ int bnxt_dev_xstats_get_op(struct rte_eth_dev *eth_dev,
 
 	for (i = 0; i < RTE_DIM(bnxt_func_stats_strings); i++) {
 		xstats[count].id = count;
-		xstats[count].value =
-			rte_le_to_cpu_64(*(uint64_t *)((char *)&func_qstats +
-					 bnxt_func_stats_strings[i].offset));
+		xstats[count].value = *(uint64_t *)((char *)&func_qstats +
+					 bnxt_func_stats_strings[i].offset);
 		count++;
 	}
 
-
 	for (i = 0; i < rx_port_stats_ext_cnt; i++) {
 		uint64_t *rx_stats_ext = (uint64_t *)bp->hw_rx_port_stats_ext;
 
@@ -820,6 +939,8 @@ int bnxt_dev_xstats_reset_op(struct rte_eth_dev *eth_dev)
 		PMD_DRV_LOG(ERR, "Failed to reset xstats: %s\n",
 			    strerror(-ret));
 
+	bnxt_clear_prev_stat(bp);
+
 	return ret;
 }
 
diff --git a/drivers/net/bnxt/bnxt_txq.c b/drivers/net/bnxt/bnxt_txq.c
index 99a31ce..795056e 100644
--- a/drivers/net/bnxt/bnxt_txq.c
+++ b/drivers/net/bnxt/bnxt_txq.c
@@ -149,8 +149,8 @@ int bnxt_tx_queue_setup_op(struct rte_eth_dev *eth_dev,
 	txq->port_id = eth_dev->data->port_id;
 
 	/* Allocate TX ring hardware descriptors */
-	if (bnxt_alloc_rings(bp, queue_idx, txq, NULL, txq->cp_ring, NULL,
-			     "txr")) {
+	if (bnxt_alloc_rings(bp, socket_id, queue_idx, txq, NULL, txq->cp_ring,
+			     NULL, "txr")) {
 		PMD_DRV_LOG(ERR, "ring_dma_zone_reserve for tx_ring failed!");
 		rc = -ENOMEM;
 		goto err;
diff --git a/drivers/net/bnxt/bnxt_txr.c b/drivers/net/bnxt/bnxt_txr.c
index 0cf6ece..ad78fa1 100644
--- a/drivers/net/bnxt/bnxt_txr.c
+++ b/drivers/net/bnxt/bnxt_txr.c
@@ -427,30 +427,26 @@ static void bnxt_tx_cmp(struct bnxt_tx_queue *txq, int nr_pkts)
 
 static int bnxt_handle_tx_cp(struct bnxt_tx_queue *txq)
 {
+	uint32_t nb_tx_pkts = 0, cons, ring_mask, opaque;
 	struct bnxt_cp_ring_info *cpr = txq->cp_ring;
 	uint32_t raw_cons = cpr->cp_raw_cons;
-	uint32_t cons;
-	uint32_t nb_tx_pkts = 0;
+	struct bnxt_ring *cp_ring_struct;
 	struct tx_cmpl *txcmp;
-	struct cmpl_base *cp_desc_ring = cpr->cp_desc_ring;
-	struct bnxt_ring *cp_ring_struct = cpr->cp_ring_struct;
-	uint32_t ring_mask = cp_ring_struct->ring_mask;
-	uint32_t opaque = 0;
 
 	if (bnxt_tx_bds_in_hw(txq) < txq->tx_free_thresh)
 		return 0;
 
+	cp_ring_struct = cpr->cp_ring_struct;
+	ring_mask = cp_ring_struct->ring_mask;
+
 	do {
 		cons = RING_CMPL(ring_mask, raw_cons);
 		txcmp = (struct tx_cmpl *)&cpr->cp_desc_ring[cons];
-		rte_prefetch_non_temporal(&cp_desc_ring[(cons + 2) &
-							ring_mask]);
 
-		if (!CMPL_VALID(txcmp, cpr->valid))
+		if (!bnxt_cpr_cmp_valid(txcmp, raw_cons, ring_mask + 1))
 			break;
-		opaque = rte_cpu_to_le_32(txcmp->opaque);
-		NEXT_CMPL(cpr, cons, cpr->valid, 1);
-		rte_prefetch0(&cp_desc_ring[cons]);
+
+		opaque = rte_le_to_cpu_32(txcmp->opaque);
 
 		if (CMP_TYPE(txcmp) == TX_CMPL_TYPE_TX_L2)
 			nb_tx_pkts += opaque;
@@ -458,9 +454,11 @@ static int bnxt_handle_tx_cp(struct bnxt_tx_queue *txq)
 			RTE_LOG_DP(ERR, PMD,
 					"Unhandled CMP type %02x\n",
 					CMP_TYPE(txcmp));
-		raw_cons = cons;
+		raw_cons = NEXT_RAW_CMP(raw_cons);
 	} while (nb_tx_pkts < ring_mask);
 
+	cpr->valid = !!(raw_cons & cp_ring_struct->ring_size);
+
 	if (nb_tx_pkts) {
 		if (txq->offloads & DEV_TX_OFFLOAD_MBUF_FAST_FREE)
 			bnxt_tx_cmp_fast(txq, nb_tx_pkts);
diff --git a/drivers/net/bnxt/bnxt_vnic.c b/drivers/net/bnxt/bnxt_vnic.c
index 007f7e9..40cad0f 100644
--- a/drivers/net/bnxt/bnxt_vnic.c
+++ b/drivers/net/bnxt/bnxt_vnic.c
@@ -145,7 +145,8 @@ int bnxt_alloc_vnic_attributes(struct bnxt *bp)
 	mz = rte_memzone_lookup(mz_name);
 	if (!mz) {
 		mz = rte_memzone_reserve(mz_name,
-				entry_length * max_vnics, SOCKET_ID_ANY,
+				entry_length * max_vnics,
+				bp->eth_dev->device->numa_node,
 				RTE_MEMZONE_2MB |
 				RTE_MEMZONE_SIZE_HINT_ONLY |
 				RTE_MEMZONE_IOVA_CONTIG);
diff --git a/drivers/net/bnxt/meson.build b/drivers/net/bnxt/meson.build
index 2896337..a2fd494 100644
--- a/drivers/net/bnxt/meson.build
+++ b/drivers/net/bnxt/meson.build
@@ -74,6 +74,6 @@ sources = files('bnxt_cpr.c',
 
 if arch_subdir == 'x86'
 	sources += files('bnxt_rxtx_vec_sse.c')
-elif arch_subdir == 'arm' and host_machine.cpu_family().startswith('aarch64')
+elif arch_subdir == 'arm' and dpdk_conf.get('RTE_ARCH_64')
 	sources += files('bnxt_rxtx_vec_neon.c')
 endif
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
index ec7db3d..a1d3703 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -1794,12 +1794,13 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev,
 				!= 0)
 			return errval;
 
-		if (bond_ethdev_8023ad_flow_verify(bonded_eth_dev,
-				slave_eth_dev->data->port_id) != 0) {
+		errval = bond_ethdev_8023ad_flow_verify(bonded_eth_dev,
+				slave_eth_dev->data->port_id);
+		if (errval != 0) {
 			RTE_BOND_LOG(ERR,
-				"rte_eth_tx_queue_setup: port=%d queue_id %d, err (%d)",
-				slave_eth_dev->data->port_id, q_id, errval);
-			return -1;
+				"bond_ethdev_8023ad_flow_verify: port=%d, err (%d)",
+				slave_eth_dev->data->port_id, errval);
+			return errval;
 		}
 
 		if (internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id] != NULL)
@@ -1807,8 +1808,14 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev,
 					internals->mode4.dedicated_queues.flow[slave_eth_dev->data->port_id],
 					&flow_error);
 
-		bond_ethdev_8023ad_flow_set(bonded_eth_dev,
+		errval = bond_ethdev_8023ad_flow_set(bonded_eth_dev,
 				slave_eth_dev->data->port_id);
+		if (errval != 0) {
+			RTE_BOND_LOG(ERR,
+				"bond_ethdev_8023ad_flow_set: port=%d, err (%d)",
+				slave_eth_dev->data->port_id, errval);
+			return errval;
+		}
 	}
 
 	/* Start device */
diff --git a/drivers/net/dpaa/dpaa_flow.c b/drivers/net/dpaa/dpaa_flow.c
index a0087df..c5b5ec8 100644
--- a/drivers/net/dpaa/dpaa_flow.c
+++ b/drivers/net/dpaa/dpaa_flow.c
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: BSD-3-Clause
- * Copyright 2017-2019 NXP
+ * Copyright 2017-2019,2021 NXP
  */
 
 /* System headers */
@@ -999,6 +999,9 @@ static int dpaa_port_vsp_configure(struct dpaa_if *dpaa_intf,
 	buf_prefix_cont.pass_time_stamp = true;
 	buf_prefix_cont.pass_hash_result = false;
 	buf_prefix_cont.pass_all_other_pcdinfo = false;
+	buf_prefix_cont.manip_ext_space =
+		RTE_PKTMBUF_HEADROOM - DPAA_MBUF_HW_ANNOTATION;
+
 	ret = fm_vsp_config_buffer_prefix_content(dpaa_intf->vsp_handle[vsp_id],
 					       &buf_prefix_cont);
 	if (ret != E_OK) {
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 69198a6..3f2c979 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -1963,6 +1963,7 @@ static int ena_dev_configure(struct rte_eth_dev *dev)
 
 	if (dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS_FLAG)
 		dev->data->dev_conf.rxmode.offloads |= DEV_RX_OFFLOAD_RSS_HASH;
+	dev->data->dev_conf.txmode.offloads |= DEV_TX_OFFLOAD_MULTI_SEGS;
 
 	adapter->tx_selected_offloads = dev->data->dev_conf.txmode.offloads;
 	adapter->rx_selected_offloads = dev->data->dev_conf.rxmode.offloads;
@@ -2037,6 +2038,7 @@ static int ena_infos_get(struct rte_eth_dev *dev,
 			DEV_RX_OFFLOAD_TCP_CKSUM;
 
 	rx_feat |= DEV_RX_OFFLOAD_JUMBO_FRAME;
+	tx_feat |= DEV_TX_OFFLOAD_MULTI_SEGS;
 
 	/* Inform framework about available features */
 	dev_info->rx_offload_capa = rx_feat;
@@ -2537,7 +2539,11 @@ static int ena_xmit_mbuf(struct ena_ring *tx_ring, struct rte_mbuf *mbuf)
 	rc = ena_com_prepare_tx(tx_ring->ena_com_io_sq,	&ena_tx_ctx,
 		&nb_hw_desc);
 	if (unlikely(rc)) {
+		PMD_DRV_LOG(ERR, "Failed to prepare Tx buffers, rc: %d\n", rc);
 		++tx_ring->tx_stats.prepare_ctx_err;
+		tx_ring->adapter->reset_reason =
+		    ENA_REGS_RESET_DRIVER_INVALID_STATE;
+		tx_ring->adapter->trigger_reset = true;
 		return rc;
 	}
 
diff --git a/drivers/net/hinic/base/hinic_pmd_niccfg.h b/drivers/net/hinic/base/hinic_pmd_niccfg.h
index 04cd374..0d0a670 100644
--- a/drivers/net/hinic/base/hinic_pmd_niccfg.h
+++ b/drivers/net/hinic/base/hinic_pmd_niccfg.h
@@ -116,15 +116,6 @@ enum hinic_link_mode {
 #define HINIC_DEFAULT_RX_MODE	(HINIC_RX_MODE_UC | HINIC_RX_MODE_MC |	\
 				HINIC_RX_MODE_BC)
 
-#define HINIC_MAX_MTU_SIZE		(9600)
-#define HINIC_MIN_MTU_SIZE		(256)
-
-/* MIN_MTU + ETH_HLEN + CRC (256+14+4) */
-#define HINIC_MIN_FRAME_SIZE		274
-
-/* MAX_MTU + ETH_HLEN + CRC + VLAN(9600+14+4+4) */
-#define HINIC_MAX_JUMBO_FRAME_SIZE	(9622)
-
 #define HINIC_PORT_DISABLE		0x0
 #define HINIC_PORT_ENABLE		0x3
 
diff --git a/drivers/net/hinic/base/hinic_pmd_nicio.c b/drivers/net/hinic/base/hinic_pmd_nicio.c
index 162308b..ad5db9f 100644
--- a/drivers/net/hinic/base/hinic_pmd_nicio.c
+++ b/drivers/net/hinic/base/hinic_pmd_nicio.c
@@ -230,8 +230,8 @@ static void hinic_rq_prepare_ctxt(struct hinic_rq *rq,
 	wq_block_pfn_hi = upper_32_bits(wq_block_pfn);
 	wq_block_pfn_lo = lower_32_bits(wq_block_pfn);
 
-	/* must config as ceq enable but do not generate ceq */
-	rq_ctxt->ceq_attr = RQ_CTXT_CEQ_ATTR_SET(1, EN) |
+	/* config as ceq disable, but must set msix state disable */
+	rq_ctxt->ceq_attr = RQ_CTXT_CEQ_ATTR_SET(0, EN) |
 			    RQ_CTXT_CEQ_ATTR_SET(1, OWNER);
 
 	rq_ctxt->pi_intr_attr = RQ_CTXT_PI_SET(pi_start, IDX) |
diff --git a/drivers/net/hinic/hinic_pmd_ethdev.c b/drivers/net/hinic/hinic_pmd_ethdev.c
index f1b3ba3..1a43597 100644
--- a/drivers/net/hinic/hinic_pmd_ethdev.c
+++ b/drivers/net/hinic/hinic_pmd_ethdev.c
@@ -69,15 +69,6 @@
 
 #define HINIC_VLAN_FILTER_EN		(1U << 0)
 
-#define HINIC_MTU_TO_PKTLEN(mtu)	\
-	((mtu) + ETH_HLEN + ETH_CRC_LEN)
-
-#define HINIC_PKTLEN_TO_MTU(pktlen)	\
-	((pktlen) - (ETH_HLEN + ETH_CRC_LEN))
-
-/* The max frame size with default MTU */
-#define HINIC_ETH_MAX_LEN (RTE_ETHER_MTU + ETH_HLEN + ETH_CRC_LEN)
-
 /* lro numer limit for one packet */
 #define HINIC_LRO_WQE_NUM_DEFAULT	8
 
@@ -1617,6 +1608,9 @@ static int hinic_vlan_filter_set(struct rte_eth_dev *dev,
 	if (vlan_id > RTE_ETHER_MAX_VLAN_ID)
 		return -EINVAL;
 
+	if (vlan_id == 0)
+		return 0;
+
 	func_id = hinic_global_func_id(nic_dev->hwdev);
 
 	if (enable) {
diff --git a/drivers/net/hinic/hinic_pmd_ethdev.h b/drivers/net/hinic/hinic_pmd_ethdev.h
index c7338d8..fafccb9 100644
--- a/drivers/net/hinic/hinic_pmd_ethdev.h
+++ b/drivers/net/hinic/hinic_pmd_ethdev.h
@@ -32,6 +32,23 @@
 #define HINIC_UINT32_BIT_SIZE      (CHAR_BIT * sizeof(uint32_t))
 #define HINIC_VFTA_SIZE            (4096 / HINIC_UINT32_BIT_SIZE)
 
+#define HINIC_MAX_MTU_SIZE              9600
+#define HINIC_MIN_MTU_SIZE              256
+
+#define HINIC_VLAN_TAG_SIZE             4
+#define HINIC_ETH_OVERHEAD \
+	(RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN + HINIC_VLAN_TAG_SIZE * 2)
+
+#define HINIC_MIN_FRAME_SIZE        (HINIC_MIN_MTU_SIZE + HINIC_ETH_OVERHEAD)
+#define HINIC_MAX_JUMBO_FRAME_SIZE  (HINIC_MAX_MTU_SIZE + HINIC_ETH_OVERHEAD)
+
+#define HINIC_MTU_TO_PKTLEN(mtu)    ((mtu) + HINIC_ETH_OVERHEAD)
+
+#define HINIC_PKTLEN_TO_MTU(pktlen) ((pktlen) - HINIC_ETH_OVERHEAD)
+
+/* The max frame size with default MTU */
+#define HINIC_ETH_MAX_LEN           (RTE_ETHER_MTU + HINIC_ETH_OVERHEAD)
+
 enum hinic_dev_status {
 	HINIC_DEV_INIT,
 	HINIC_DEV_CLOSE,
diff --git a/drivers/net/hns3/hns3_cmd.c b/drivers/net/hns3/hns3_cmd.c
index c6552b4..dc65ef8 100644
--- a/drivers/net/hns3/hns3_cmd.c
+++ b/drivers/net/hns3/hns3_cmd.c
@@ -44,10 +44,12 @@ static int
 hns3_allocate_dma_mem(struct hns3_hw *hw, struct hns3_cmq_ring *ring,
 		      uint64_t size, uint32_t alignment)
 {
+	static uint64_t hns3_dma_memzone_id;
 	const struct rte_memzone *mz = NULL;
 	char z_name[RTE_MEMZONE_NAMESIZE];
 
-	snprintf(z_name, sizeof(z_name), "hns3_dma_%" PRIu64, rte_rand());
+	snprintf(z_name, sizeof(z_name), "hns3_dma_%" PRIu64,
+		__atomic_fetch_add(&hns3_dma_memzone_id, 1, __ATOMIC_RELAXED));
 	mz = rte_memzone_reserve_bounded(z_name, size, SOCKET_ID_ANY,
 					 RTE_MEMZONE_IOVA_CONTIG, alignment,
 					 RTE_PGSIZE_2M);
diff --git a/drivers/net/hns3/hns3_ethdev.c b/drivers/net/hns3/hns3_ethdev.c
index 637e613..abe31c8 100644
--- a/drivers/net/hns3/hns3_ethdev.c
+++ b/drivers/net/hns3/hns3_ethdev.c
@@ -239,6 +239,7 @@ hns3_interrupt_handler(void *param)
 	hns3_pf_disable_irq0(hw);
 
 	event_cause = hns3_check_event_cause(hns, &clearval);
+	hns3_clear_event_cause(hw, event_cause, clearval);
 	/* vector 0 interrupt is shared with reset and mailbox source events. */
 	if (event_cause == HNS3_VECTOR0_EVENT_ERR) {
 		hns3_warn(hw, "Received err interrupt");
@@ -253,7 +254,6 @@ hns3_interrupt_handler(void *param)
 	else
 		hns3_err(hw, "Received unknown event");
 
-	hns3_clear_event_cause(hw, event_cause, clearval);
 	/* Enable interrupt if it is not cause by reset */
 	hns3_pf_enable_irq0(hw);
 }
@@ -558,7 +558,8 @@ hns3_en_hw_strip_rxvtag(struct hns3_adapter *hns, bool enable)
 
 	ret = hns3_set_vlan_rx_offload_cfg(hns, &rxvlan_cfg);
 	if (ret) {
-		hns3_err(hw, "enable strip rx vtag failed, ret =%d", ret);
+		hns3_err(hw, "%s strip rx vtag failed, ret = %d.",
+				enable ? "enable" : "disable", ret);
 		return ret;
 	}
 
@@ -1665,7 +1666,6 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 	struct rte_ether_addr *oaddr;
 	char mac_str[RTE_ETHER_ADDR_FMT_SIZE];
 	bool default_addr_setted;
-	bool rm_succes = false;
 	int ret, ret_val;
 
 	/*
@@ -1685,9 +1685,10 @@ hns3_set_default_mac_addr(struct rte_eth_dev *dev,
 					      oaddr);
 			hns3_warn(hw, "Remove old uc mac address(%s) fail: %d",
 				  mac_str, ret);
-			rm_succes = false;
-		} else
-			rm_succes = true;
+
+			rte_spinlock_unlock(&hw->lock);
+			return ret;
+		}
 	}
 
 	ret = hns3_add_uc_addr_common(hw, mac_addr);
@@ -1722,16 +1723,12 @@ err_pause_addr_cfg:
 	}
 
 err_add_uc_addr:
-	if (rm_succes) {
-		ret_val = hns3_add_uc_addr_common(hw, oaddr);
-		if (ret_val) {
-			rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE,
-					      oaddr);
-			hns3_warn(hw,
-				  "Failed to restore old uc mac addr(%s): %d",
+	ret_val = hns3_add_uc_addr_common(hw, oaddr);
+	if (ret_val) {
+		rte_ether_format_addr(mac_str, RTE_ETHER_ADDR_FMT_SIZE, oaddr);
+		hns3_warn(hw, "Failed to restore old uc mac addr(%s): %d",
 				  mac_str, ret_val);
-			hw->mac.default_addr_setted = false;
-		}
+		hw->mac.default_addr_setted = false;
 	}
 	rte_spinlock_unlock(&hw->lock);
 
@@ -2376,13 +2373,11 @@ hns3_dev_configure(struct rte_eth_dev *dev)
 	 * work as usual. But these fake queues are imperceptible, and can not
 	 * be used by upper applications.
 	 */
-	if (!hns3_dev_indep_txrx_supported(hw)) {
-		ret = hns3_set_fake_rx_or_tx_queues(dev, nb_rx_q, nb_tx_q);
-		if (ret) {
-			hns3_err(hw, "fail to set Rx/Tx fake queues, ret = %d.",
-				 ret);
-			return ret;
-		}
+	ret = hns3_set_fake_rx_or_tx_queues(dev, nb_rx_q, nb_tx_q);
+	if (ret) {
+		hns3_err(hw, "fail to set Rx/Tx fake queues, ret = %d.", ret);
+		hw->cfg_max_queues = 0;
+		return ret;
 	}
 
 	hw->adapter_state = HNS3_NIC_CONFIGURING;
@@ -2441,6 +2436,7 @@ hns3_dev_configure(struct rte_eth_dev *dev)
 	return 0;
 
 cfg_err:
+	hw->cfg_max_queues = 0;
 	(void)hns3_set_fake_rx_or_tx_queues(dev, 0, 0);
 	hw->adapter_state = HNS3_NIC_INITIALIZED;
 
@@ -4856,6 +4852,7 @@ hns3_uninit_pf(struct rte_eth_dev *eth_dev)
 	hns3_rss_uninit(hns);
 	(void)hns3_config_gro(hw, false);
 	hns3_promisc_uninit(hw);
+	hns3_flow_uninit(eth_dev);
 	hns3_fdir_filter_uninit(hns);
 	(void)hns3_firmware_compat_config(hw, false);
 	hns3_uninit_umv_space(hw);
@@ -5169,7 +5166,7 @@ hns3_dev_stop(struct rte_eth_dev *dev)
 	/* Disable datapath on secondary process. */
 	hns3_mp_req_stop_rxtx(dev);
 	/* Prevent crashes when queues are still in use. */
-	rte_delay_ms(hw->tqps_num);
+	rte_delay_ms(hw->cfg_max_queues);
 
 	rte_spinlock_lock(&hw->lock);
 	if (rte_atomic16_read(&hw->reset.resetting) == 0) {
@@ -5192,11 +5189,8 @@ hns3_dev_close(struct rte_eth_dev *eth_dev)
 	struct hns3_hw *hw = &hns->hw;
 	int ret = 0;
 
-	if (rte_eal_process_type() != RTE_PROC_PRIMARY) {
-		rte_free(eth_dev->process_private);
-		eth_dev->process_private = NULL;
+	if (rte_eal_process_type() != RTE_PROC_PRIMARY)
 		return 0;
-	}
 
 	if (hw->adapter_state == HNS3_NIC_STARTED)
 		ret = hns3_dev_stop(eth_dev);
@@ -5211,8 +5205,6 @@ hns3_dev_close(struct rte_eth_dev *eth_dev)
 	hns3_uninit_pf(eth_dev);
 	hns3_free_all_queues(eth_dev);
 	rte_free(hw->reset.wait_data);
-	rte_free(eth_dev->process_private);
-	eth_dev->process_private = NULL;
 	hns3_mp_uninit_primary();
 	hns3_warn(hw, "Close port %u finished", hw->data->port_id);
 
@@ -5657,7 +5649,7 @@ hns3_stop_service(struct hns3_adapter *hns)
 	rte_wmb();
 	/* Disable datapath on secondary process. */
 	hns3_mp_req_stop_rxtx(eth_dev);
-	rte_delay_ms(hw->tqps_num);