Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

drivers/net/ethernet/mellanox/mlx5/core/main.c
  b33886971dbc ("net/mlx5: Initialize flow steering during driver probe")
  40379a0084c2 ("net/mlx5_fpga: Drop INNOVA TLS support")
  f2b41b32cde8 ("net/mlx5: Remove ipsec_ops function table")
https://lore.kernel.org/all/20220519040345.6yrjromcdistu7vh@sx1/
  16d42d313350 ("net/mlx5: Drain fw_reset when removing device")
  8324a02c342a ("net/mlx5: Add exit route when waiting for FW")
https://lore.kernel.org/all/20220519114119.060ce014@canb.auug.org.au/

tools/testing/selftests/net/mptcp/mptcp_join.sh
  e274f7154008 ("selftests: mptcp: add subflow limits test-cases")
  b6e074e171bc ("selftests: mptcp: add infinite map testcase")
  5ac1d2d63451 ("selftests: mptcp: Add tests for userspace PM type")
https://lore.kernel.org/all/20220516111918.366d747f@canb.auug.org.au/

net/mptcp/options.c
  ba2c89e0ea74 ("mptcp: fix checksum byte order")
  1e39e5a32ad7 ("mptcp: infinite mapping sending")
  ea66758c1795 ("tcp: allow MPTCP to update the announced window")
https://lore.kernel.org/all/20220519115146.751c3a37@canb.auug.org.au/

net/mptcp/pm.c
  95d686517884 ("mptcp: fix subflow accounting on close")
  4d25247d3ae4 ("mptcp: bypass in-kernel PM restrictions for non-kernel PMs")
https://lore.kernel.org/all/20220516111435.72f35dca@canb.auug.org.au/

net/mptcp/subflow.c
  ae66fb2ba6c3 ("mptcp: Do TCP fallback on early DSS checksum failure")
  0348c690ed37 ("mptcp: add the fallback check")
  f8d4bcacff3b ("mptcp: infinite mapping receiving")
https://lore.kernel.org/all/20220519115837.380bb8d4@canb.auug.org.au/

Signed-off-by: Jakub Kicinski <kuba@kernel.org>

14 files changed, 246 insertions, 123 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/ct_fs_smfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/ct_fs_smfs.c
index b979826f3f6c..2b80fe73549d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/ct_fs_smfs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/ct_fs_smfs.c
@@ -23,7 +23,7 @@ struct mlx5_ct_fs_smfs_matcher {
 };
 
 struct mlx5_ct_fs_smfs_matchers {
-	struct mlx5_ct_fs_smfs_matcher smfs_matchers[4];
+	struct mlx5_ct_fs_smfs_matcher smfs_matchers[6];
 	struct list_head used;
 };
 
@@ -44,7 +44,8 @@ struct mlx5_ct_fs_smfs_rule {
 };
 
 static inline void
-mlx5_ct_fs_smfs_fill_mask(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec, bool ipv4, bool tcp)
+mlx5_ct_fs_smfs_fill_mask(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec, bool ipv4, bool tcp,
+			  bool gre)
 {
 	void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers);
 
@@ -77,7 +78,7 @@ mlx5_ct_fs_smfs_fill_mask(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec, bo
 		MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, tcp_dport);
 		MLX5_SET(fte_match_set_lyr_2_4, headers_c, tcp_flags,
 			 ntohs(MLX5_CT_TCP_FLAGS_MASK));
-	} else {
+	} else if (!gre) {
 		MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, udp_sport);
 		MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, udp_dport);
 	}
@@ -87,7 +88,7 @@ mlx5_ct_fs_smfs_fill_mask(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec, bo
 
 static struct mlx5dr_matcher *
 mlx5_ct_fs_smfs_matcher_create(struct mlx5_ct_fs *fs, struct mlx5dr_table *tbl, bool ipv4,
-			       bool tcp, u32 priority)
+			       bool tcp, bool gre, u32 priority)
 {
 	struct mlx5dr_matcher *dr_matcher;
 	struct mlx5_flow_spec *spec;
@@ -96,7 +97,7 @@ mlx5_ct_fs_smfs_matcher_create(struct mlx5_ct_fs *fs, struct mlx5dr_table *tbl,
 	if (!spec)
 		return ERR_PTR(-ENOMEM);
 
-	mlx5_ct_fs_smfs_fill_mask(fs, spec, ipv4, tcp);
+	mlx5_ct_fs_smfs_fill_mask(fs, spec, ipv4, tcp, gre);
 	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2 | MLX5_MATCH_OUTER_HEADERS;
 
 	dr_matcher = mlx5_smfs_matcher_create(tbl, priority, spec);
@@ -108,7 +109,7 @@ mlx5_ct_fs_smfs_matcher_create(struct mlx5_ct_fs *fs, struct mlx5dr_table *tbl,
 }
 
 static struct mlx5_ct_fs_smfs_matcher *
-mlx5_ct_fs_smfs_matcher_get(struct mlx5_ct_fs *fs, bool nat, bool ipv4, bool tcp)
+mlx5_ct_fs_smfs_matcher_get(struct mlx5_ct_fs *fs, bool nat, bool ipv4, bool tcp, bool gre)
 {
 	struct mlx5_ct_fs_smfs *fs_smfs = mlx5_ct_fs_priv(fs);
 	struct mlx5_ct_fs_smfs_matcher *m, *smfs_matcher;
@@ -119,7 +120,7 @@ mlx5_ct_fs_smfs_matcher_get(struct mlx5_ct_fs *fs, bool nat, bool ipv4, bool tcp
 	int prio;
 
 	matchers = nat ? &fs_smfs->matchers_nat : &fs_smfs->matchers;
-	smfs_matcher = &matchers->smfs_matchers[ipv4 * 2 + tcp];
+	smfs_matcher = &matchers->smfs_matchers[ipv4 * 3 + tcp * 2 + gre];
 
 	if (refcount_inc_not_zero(&smfs_matcher->ref))
 		return smfs_matcher;
@@ -145,11 +146,11 @@ mlx5_ct_fs_smfs_matcher_get(struct mlx5_ct_fs *fs, bool nat, bool ipv4, bool tcp
 	}
 
 	tbl = nat ? fs_smfs->ct_nat_tbl : fs_smfs->ct_tbl;
-	dr_matcher = mlx5_ct_fs_smfs_matcher_create(fs, tbl, ipv4, tcp, prio);
+	dr_matcher = mlx5_ct_fs_smfs_matcher_create(fs, tbl, ipv4, tcp, gre, prio);
 	if (IS_ERR(dr_matcher)) {
 		netdev_warn(fs->netdev,
-			    "ct_fs_smfs: failed to create matcher (nat %d, ipv4 %d, tcp %d), err: %ld\n",
-			    nat, ipv4, tcp, PTR_ERR(dr_matcher));
+			    "ct_fs_smfs: failed to create matcher (nat %d, ipv4 %d, tcp %d, gre %d), err: %ld\n",
+			    nat, ipv4, tcp, gre, PTR_ERR(dr_matcher));
 
 		smfs_matcher = ERR_CAST(dr_matcher);
 		goto out_unlock;
@@ -222,16 +223,17 @@ mlx5_ct_fs_smfs_destroy(struct mlx5_ct_fs *fs)
 static inline bool
 mlx5_tc_ct_valid_used_dissector_keys(const u32 used_keys)
 {
-#define DISSECTOR_BIT(name) BIT(FLOW_DISSECTOR_KEY_ ## name)
-	const u32 basic_keys = DISSECTOR_BIT(BASIC) | DISSECTOR_BIT(CONTROL) |
-			       DISSECTOR_BIT(PORTS) | DISSECTOR_BIT(META);
-	const u32 ipv4_tcp = basic_keys | DISSECTOR_BIT(IPV4_ADDRS) | DISSECTOR_BIT(TCP);
-	const u32 ipv4_udp = basic_keys | DISSECTOR_BIT(IPV4_ADDRS);
-	const u32 ipv6_tcp = basic_keys | DISSECTOR_BIT(IPV6_ADDRS) | DISSECTOR_BIT(TCP);
-	const u32 ipv6_udp = basic_keys | DISSECTOR_BIT(IPV6_ADDRS);
+#define DISS_BIT(name) BIT(FLOW_DISSECTOR_KEY_ ## name)
+	const u32 basic_keys = DISS_BIT(BASIC) | DISS_BIT(CONTROL) | DISS_BIT(META);
+	const u32 ipv4_tcp = basic_keys | DISS_BIT(IPV4_ADDRS) | DISS_BIT(PORTS) | DISS_BIT(TCP);
+	const u32 ipv6_tcp = basic_keys | DISS_BIT(IPV6_ADDRS) | DISS_BIT(PORTS) | DISS_BIT(TCP);
+	const u32 ipv4_udp = basic_keys | DISS_BIT(IPV4_ADDRS) | DISS_BIT(PORTS);
+	const u32 ipv6_udp = basic_keys | DISS_BIT(IPV6_ADDRS) | DISS_BIT(PORTS);
+	const u32 ipv4_gre = basic_keys | DISS_BIT(IPV4_ADDRS);
+	const u32 ipv6_gre = basic_keys | DISS_BIT(IPV6_ADDRS);
 
 	return (used_keys == ipv4_tcp || used_keys == ipv4_udp || used_keys == ipv6_tcp ||
-		used_keys == ipv6_udp);
+		used_keys == ipv6_udp || used_keys == ipv4_gre || used_keys == ipv6_gre);
 }
 
 static bool
@@ -254,20 +256,24 @@ mlx5_ct_fs_smfs_ct_validate_flow_rule(struct mlx5_ct_fs *fs, struct flow_rule *f
 	flow_rule_match_control(flow_rule, &control);
 	flow_rule_match_ipv4_addrs(flow_rule, &ipv4_addrs);
 	flow_rule_match_ipv6_addrs(flow_rule, &ipv6_addrs);
-	flow_rule_match_ports(flow_rule, &ports);
-	flow_rule_match_tcp(flow_rule, &tcp);
+	if (basic.key->ip_proto != IPPROTO_GRE)
+		flow_rule_match_ports(flow_rule, &ports);
+	if (basic.key->ip_proto == IPPROTO_TCP)
+		flow_rule_match_tcp(flow_rule, &tcp);
 
 	if (basic.mask->n_proto != htons(0xFFFF) ||
 	    (basic.key->n_proto != htons(ETH_P_IP) && basic.key->n_proto != htons(ETH_P_IPV6)) ||
 	    basic.mask->ip_proto != 0xFF ||
-	    (basic.key->ip_proto != IPPROTO_UDP && basic.key->ip_proto != IPPROTO_TCP)) {
+	    (basic.key->ip_proto != IPPROTO_UDP && basic.key->ip_proto != IPPROTO_TCP &&
+	     basic.key->ip_proto != IPPROTO_GRE)) {
 		ct_dbg("rule uses unexpected basic match (n_proto 0x%04x/0x%04x, ip_proto 0x%02x/0x%02x)",
 		       ntohs(basic.key->n_proto), ntohs(basic.mask->n_proto),
 		       basic.key->ip_proto, basic.mask->ip_proto);
 		return false;
 	}
 
-	if (ports.mask->src != htons(0xFFFF) || ports.mask->dst != htons(0xFFFF)) {
+	if (basic.key->ip_proto != IPPROTO_GRE &&
+	    (ports.mask->src != htons(0xFFFF) || ports.mask->dst != htons(0xFFFF))) {
 		ct_dbg("rule uses ports match (src 0x%04x, dst 0x%04x)",
 		       ports.mask->src, ports.mask->dst);
 		return false;
@@ -291,7 +297,7 @@ mlx5_ct_fs_smfs_ct_rule_add(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec,
 	struct mlx5dr_action *actions[5];
 	struct mlx5dr_rule *rule;
 	int num_actions = 0, err;
-	bool nat, tcp, ipv4;
+	bool nat, tcp, ipv4, gre;
 
 	if (!mlx5_ct_fs_smfs_ct_validate_flow_rule(fs, flow_rule))
 		return ERR_PTR(-EOPNOTSUPP);
@@ -314,15 +320,17 @@ mlx5_ct_fs_smfs_ct_rule_add(struct mlx5_ct_fs *fs, struct mlx5_flow_spec *spec,
 	ipv4 = mlx5e_tc_get_ip_version(spec, true) == 4;
 	tcp = MLX5_GET(fte_match_param, spec->match_value,
 		       outer_headers.ip_protocol) == IPPROTO_TCP;
+	gre = MLX5_GET(fte_match_param, spec->match_value,
+		       outer_headers.ip_protocol) == IPPROTO_GRE;
 
-	smfs_matcher = mlx5_ct_fs_smfs_matcher_get(fs, nat, ipv4, tcp);
+	smfs_matcher = mlx5_ct_fs_smfs_matcher_get(fs, nat, ipv4, tcp, gre);
 	if (IS_ERR(smfs_matcher)) {
 		err = PTR_ERR(smfs_matcher);
 		goto err_matcher;
 	}
 
 	rule = mlx5_smfs_rule_create(smfs_matcher->dr_matcher, spec, num_actions, actions,
-				     MLX5_FLOW_CONTEXT_FLOW_SOURCE_ANY_VPORT);
+				     spec->flow_context.flow_source);
 	if (!rule) {
 		err = -EINVAL;
 		goto err_create;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c
index a55b066746cb..857840ab1e91 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/trap.c
@@ -14,19 +14,26 @@ static int mlx5e_trap_napi_poll(struct napi_struct *napi, int budget)
 	bool busy = false;
 	int work_done = 0;
 
+	rcu_read_lock();
+
 	ch_stats->poll++;
 
 	work_done = mlx5e_poll_rx_cq(&rq->cq, budget);
 	busy |= work_done == budget;
 	busy |= rq->post_wqes(rq);
 
-	if (busy)
-		return budget;
+	if (busy) {
+		work_done = budget;
+		goto out;
+	}
 
 	if (unlikely(!napi_complete_done(napi, work_done)))
-		return work_done;
+		goto out;
 
 	mlx5e_cq_arm(&rq->cq);
+
+out:
+	rcu_read_unlock();
 	return work_done;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6816a024db8a..05c015515cce 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3858,6 +3858,10 @@ static netdev_features_t mlx5e_fix_uplink_rep_features(struct net_device *netdev
 	if (netdev->features & NETIF_F_NTUPLE)
 		netdev_warn(netdev, "Disabling ntuple, not supported in switchdev mode\n");
 
+	features &= ~NETIF_F_GRO_HW;
+	if (netdev->features & NETIF_F_GRO_HW)
+		netdev_warn(netdev, "Disabling HW_GRO, not supported in switchdev mode\n");
+
 	return features;
 }
 
@@ -3890,6 +3894,25 @@ static netdev_features_t mlx5e_fix_features(struct net_device *netdev,
 		}
 	}
 
+	if (params->xdp_prog) {
+		if (features & NETIF_F_LRO) {
+			netdev_warn(netdev, "LRO is incompatible with XDP\n");
+			features &= ~NETIF_F_LRO;
+		}
+		if (features & NETIF_F_GRO_HW) {
+			netdev_warn(netdev, "HW GRO is incompatible with XDP\n");
+			features &= ~NETIF_F_GRO_HW;
+		}
+	}
+
+	if (priv->xsk.refcnt) {
+		if (features & NETIF_F_GRO_HW) {
+			netdev_warn(netdev, "HW GRO is incompatible with AF_XDP (%u XSKs are active)\n",
+				    priv->xsk.refcnt);
+			features &= ~NETIF_F_GRO_HW;
+		}
+	}
+
 	if (MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)) {
 		features &= ~NETIF_F_RXHASH;
 		if (netdev->features & NETIF_F_RXHASH)
@@ -4839,10 +4862,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 	netdev->hw_features      |= NETIF_F_HW_VLAN_CTAG_FILTER;
 	netdev->hw_features      |= NETIF_F_HW_VLAN_STAG_TX;
 
-	if (!!MLX5_CAP_GEN(mdev, shampo) &&
-	    mlx5e_check_fragmented_striding_rq_cap(mdev))
-		netdev->hw_features    |= NETIF_F_GRO_HW;
-
 	if (mlx5e_tunnel_any_tx_proto_supported(mdev)) {
 		netdev->hw_enc_features |= NETIF_F_HW_CSUM;
 		netdev->hw_enc_features |= NETIF_F_TSO;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
index fb8175672478..84caffe4c278 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c
@@ -2673,28 +2673,6 @@ static void cleanup_root_ns(struct mlx5_flow_root_namespace *root_ns)
 	clean_tree(&root_ns->ns.node);
 }
 
-void mlx5_cleanup_fs(struct mlx5_core_dev *dev)
-{
-	struct mlx5_flow_steering *steering = dev->priv.steering;
-
-	cleanup_root_ns(steering->root_ns);
-	cleanup_root_ns(steering->fdb_root_ns);
-	steering->fdb_root_ns = NULL;
-	kfree(steering->fdb_sub_ns);
-	steering->fdb_sub_ns = NULL;
-	cleanup_root_ns(steering->port_sel_root_ns);
-	cleanup_root_ns(steering->sniffer_rx_root_ns);
-	cleanup_root_ns(steering->sniffer_tx_root_ns);
-	cleanup_root_ns(steering->rdma_rx_root_ns);
-	cleanup_root_ns(steering->rdma_tx_root_ns);
-	cleanup_root_ns(steering->egress_root_ns);
-	mlx5_cleanup_fc_stats(dev);
-	kmem_cache_destroy(steering->ftes_cache);
-	kmem_cache_destroy(steering->fgs_cache);
-	mlx5_ft_pool_destroy(dev);
-	kfree(steering);
-}
-
 static int init_sniffer_tx_root_ns(struct mlx5_flow_steering *steering)
 {
 	struct fs_prio *prio;
@@ -3096,42 +3074,27 @@ cleanup:
 	return err;
 }
 
-int mlx5_init_fs(struct mlx5_core_dev *dev)
+void mlx5_fs_core_cleanup(struct mlx5_core_dev *dev)
 {
-	struct mlx5_flow_steering *steering;
-	int err = 0;
-
-	err = mlx5_init_fc_stats(dev);
-	if (err)
-		return err;
-
-	err = mlx5_ft_pool_init(dev);
-	if (err)
-		return err;
-
-	steering = kzalloc(sizeof(*steering), GFP_KERNEL);
-	if (!steering) {
-		err = -ENOMEM;
-		goto err;
-	}
-
-	steering->dev = dev;
-	dev->priv.steering = steering;
+	struct mlx5_flow_steering *steering = dev->priv.steering;
 
-	if (mlx5_fs_dr_is_supported(dev))
-		steering->mode = MLX5_FLOW_STEERING_MODE_SMFS;
-	else
-		steering->mode = MLX5_FLOW_STEERING_MODE_DMFS;
+	cleanup_root_ns(steering->root_ns);
+	cleanup_root_ns(steering->fdb_root_ns);
+	steering->fdb_root_ns = NULL;
+	kfree(steering->fdb_sub_ns);
+	steering->fdb_sub_ns = NULL;
+	cleanup_root_ns(steering->port_sel_root_ns);
+	cleanup_root_ns(steering->sniffer_rx_root_ns);
+	cleanup_root_ns(steering->sniffer_tx_root_ns);
+	cleanup_root_ns(steering->rdma_rx_root_ns);
+	cleanup_root_ns(steering->rdma_tx_root_ns);
+	cleanup_root_ns(steering->egress_root_ns);
+}
 
-	steering->fgs_cache = kmem_cache_create("mlx5_fs_fgs",
-						sizeof(struct mlx5_flow_group), 0,
-						0, NULL);
-	steering->ftes_cache = kmem_cache_create("mlx5_fs_ftes", sizeof(struct fs_fte), 0,
-						 0, NULL);
-	if (!steering->ftes_cache || !steering->fgs_cache) {
-		err = -ENOMEM;
-		goto err;
-	}
+int mlx5_fs_core_init(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_steering *steering = dev->priv.steering;
+	int err = 0;
 
 	if ((((MLX5_CAP_GEN(dev, port_type) == MLX5_CAP_PORT_TYPE_ETH) &&
 	      (MLX5_CAP_GEN(dev, nic_flow_table))) ||
@@ -3189,8 +3152,64 @@ int mlx5_init_fs(struct mlx5_core_dev *dev)
 	}
 
 	return 0;
+
+err:
+	mlx5_fs_core_cleanup(dev);
+	return err;
+}
+
+void mlx5_fs_core_free(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_steering *steering = dev->priv.steering;
+
+	kmem_cache_destroy(steering->ftes_cache);
+	kmem_cache_destroy(steering->fgs_cache);
+	kfree(steering);
+	mlx5_ft_pool_destroy(dev);
+	mlx5_cleanup_fc_stats(dev);
+}
+
+int mlx5_fs_core_alloc(struct mlx5_core_dev *dev)
+{
+	struct mlx5_flow_steering *steering;
+	int err = 0;
+
+	err = mlx5_init_fc_stats(dev);
+	if (err)
+		return err;
+
+	err = mlx5_ft_pool_init(dev);
+	if (err)
+		goto err;
+
+	steering = kzalloc(sizeof(*steering), GFP_KERNEL);
+	if (!steering) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	steering->dev = dev;
+	dev->priv.steering = steering;
+
+	if (mlx5_fs_dr_is_supported(dev))
+		steering->mode = MLX5_FLOW_STEERING_MODE_SMFS;
+	else
+		steering->mode = MLX5_FLOW_STEERING_MODE_DMFS;
+
+	steering->fgs_cache = kmem_cache_create("mlx5_fs_fgs",
+						sizeof(struct mlx5_flow_group), 0,
+						0, NULL);
+	steering->ftes_cache = kmem_cache_create("mlx5_fs_ftes", sizeof(struct fs_fte), 0,
+						 0, NULL);
+	if (!steering->ftes_cache || !steering->fgs_cache) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	return 0;
+
 err:
-	mlx5_cleanup_fs(dev);
+	mlx5_fs_core_free(dev);
 	return err;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
index 67cad7a6d836..3af50fd04d28 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h
@@ -299,8 +299,10 @@ int mlx5_flow_namespace_set_peer(struct mlx5_flow_root_namespace *ns,
 int mlx5_flow_namespace_set_mode(struct mlx5_flow_namespace *ns,
 				 enum mlx5_flow_steering_mode mode);
 
-int mlx5_init_fs(struct mlx5_core_dev *dev);
-void mlx5_cleanup_fs(struct mlx5_core_dev *dev);
+int mlx5_fs_core_alloc(struct mlx5_core_dev *dev);
+void mlx5_fs_core_free(struct mlx5_core_dev *dev);
+int mlx5_fs_core_init(struct mlx5_core_dev *dev);
+void mlx5_fs_core_cleanup(struct mlx5_core_dev *dev);
 
 int mlx5_fs_egress_acls_init(struct mlx5_core_dev *dev, int total_vports);
 void mlx5_fs_egress_acls_cleanup(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
index 84df0d56a2b6..052af4901c0b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c
@@ -8,7 +8,8 @@
 enum {
 	MLX5_FW_RESET_FLAGS_RESET_REQUESTED,
 	MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST,
-	MLX5_FW_RESET_FLAGS_PENDING_COMP
+	MLX5_FW_RESET_FLAGS_PENDING_COMP,
+	MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS
 };
 
 struct mlx5_fw_reset {
@@ -208,7 +209,10 @@ static void poll_sync_reset(struct timer_list *t)
 
 	if (fatal_error) {
 		mlx5_core_warn(dev, "Got Device Reset\n");
-		queue_work(fw_reset->wq, &fw_reset->reset_reload_work);
+		if (!test_bit(MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS, &fw_reset->reset_flags))
+			queue_work(fw_reset->wq, &fw_reset->reset_reload_work);
+		else
+			mlx5_core_err(dev, "Device is being removed, Drop new reset work\n");
 		return;
 	}
 
@@ -433,9 +437,12 @@ static int fw_reset_event_notifier(struct notifier_block *nb, unsigned long acti
 	struct mlx5_fw_reset *fw_reset = mlx5_nb_cof(nb, struct mlx5_fw_reset, nb);
 	struct mlx5_eqe *eqe = data;
 
+	if (test_bit(MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS, &fw_reset->reset_flags))
+		return NOTIFY_DONE;
+
 	switch (eqe->sub_type) {
 	case MLX5_GENERAL_SUBTYPE_FW_LIVE_PATCH_EVENT:
-			queue_work(fw_reset->wq, &fw_reset->fw_live_patch_work);
+		queue_work(fw_reset->wq, &fw_reset->fw_live_patch_work);
 		break;
 	case MLX5_GENERAL_SUBTYPE_PCI_SYNC_FOR_FW_UPDATE_EVENT:
 		mlx5_sync_reset_events_handle(fw_reset, eqe);
@@ -479,6 +486,18 @@ void mlx5_fw_reset_events_stop(struct mlx5_core_dev *dev)
 	mlx5_eq_notifier_unregister(dev, &dev->priv.fw_reset->nb);
 }
 
+void mlx5_drain_fw_reset(struct mlx5_core_dev *dev)
+{
+	struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset;
+
+	set_bit(MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS, &fw_reset->reset_flags);
+	cancel_work_sync(&fw_reset->fw_live_patch_work);
+	cancel_work_sync(&fw_reset->reset_request_work);
+	cancel_work_sync(&fw_reset->reset_reload_work);
+	cancel_work_sync(&fw_reset->reset_now_work);
+	cancel_work_sync(&fw_reset->reset_abort_work);
+}
+
 int mlx5_fw_reset_init(struct mlx5_core_dev *dev)
 {
 	struct mlx5_fw_reset *fw_reset = kzalloc(sizeof(*fw_reset), GFP_KERNEL);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h
index 694fc7cb2684..dc141c7e641a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h
@@ -16,6 +16,7 @@ int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev);
 int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev);
 void mlx5_fw_reset_events_start(struct mlx5_core_dev *dev);
 void mlx5_fw_reset_events_stop(struct mlx5_core_dev *dev);
+void mlx5_drain_fw_reset(struct mlx5_core_dev *dev);
 int mlx5_fw_reset_init(struct mlx5_core_dev *dev);
 void mlx5_fw_reset_cleanup(struct mlx5_core_dev *dev);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 87f1552b5d73..fbd4dd76a19f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -936,6 +936,12 @@ static int mlx5_init_once(struct mlx5_core_dev *dev)
 		goto err_sf_table_cleanup;
 	}
 
+	err = mlx5_fs_core_alloc(dev);
+	if (err) {
+		mlx5_core_err(dev, "Failed to alloc flow steering\n");
+		goto err_fs;
+	}
+
 	dev->dm = mlx5_dm_create(dev);
 	if (IS_ERR(dev->dm))
 		mlx5_core_warn(dev, "Failed to init device memory%d\n", err);
@@ -946,6 +952,8 @@ static int mlx5_init_once(struct mlx5_core_dev *dev)
 
 	return 0;
 
+err_fs:
+	mlx5_sf_table_cleanup(dev);
 err_sf_table_cleanup:
 	mlx5_sf_hw_table_cleanup(dev);
 err_sf_hw_table_cleanup:
@@ -983,6 +991,7 @@ static void mlx5_cleanup_once(struct mlx5_core_dev *dev)
 	mlx5_hv_vhca_destroy(dev->hv_vhca);
 	mlx5_fw_tracer_destroy(dev->tracer);
 	mlx5_dm_cleanup(dev);
+	mlx5_fs_core_free(dev);
 	mlx5_sf_table_cleanup(dev);
 	mlx5_sf_hw_table_cleanup(dev);
 	mlx5_vhca_event_cleanup(dev);
@@ -1181,7 +1190,7 @@ static int mlx5_load(struct mlx5_core_dev *dev)
 		goto err_fpga_start;
 	}
 
-	err = mlx5_init_fs(dev);
+	err = mlx5_fs_core_init(dev);
 	if (err) {
 		mlx5_core_err(dev, "Failed to init flow steering\n");
 		goto err_fs;
@@ -1226,7 +1235,7 @@ err_ec:
 err_vhca:
 	mlx5_vhca_event_stop(dev);
 err_set_hca:
-	mlx5_cleanup_fs(dev);
+	mlx5_fs_core_cleanup(dev);
 err_fs:
 	mlx5_fpga_device_stop(dev);
 err_fpga_start:
@@ -1252,7 +1261,7 @@ static void mlx5_unload(struct mlx5_core_dev *dev)
 	mlx5_ec_cleanup(dev);
 	mlx5_sf_hw_table_destroy(dev);
 	mlx5_vhca_event_stop(dev);
-	mlx5_cleanup_fs(dev);
+	mlx5_fs_core_cleanup(dev);
 	mlx5_fpga_device_stop(dev);
 	mlx5_rsc_dump_cleanup(dev);
 	mlx5_hv_vhca_cleanup(dev->hv_vhca);
@@ -1608,6 +1617,10 @@ static void remove_one(struct pci_dev *pdev)
 	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
 	struct devlink *devlink = priv_to_devlink(dev);
 
+	/* mlx5_drain_fw_reset() is using devlink APIs. Hence, we must drain
+	 * fw_reset before unregistering the devlink.
+	 */
+	mlx5_drain_fw_reset(dev);
 	set_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state);
 	devlink_unregister(devlink);
 	mlx5_sriov_disable(pdev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 850937cd8bf9..1383550f44c1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -530,6 +530,37 @@ static int dr_action_handle_cs_recalc(struct mlx5dr_domain *dmn,
 	return 0;
 }
 
+static void dr_action_modify_ttl_adjust(struct mlx5dr_domain *dmn,
+					struct mlx5dr_ste_actions_attr *attr,
+					bool rx_rule,
+					bool *recalc_cs_required)
+{
+	*recalc_cs_required = false;
+
+	/* if device supports csum recalculation - no adjustment needed */
+	if (mlx5dr_ste_supp_ttl_cs_recalc(&dmn->info.caps))
+		return;
+
+	/* no adjustment needed on TX rules */
+	if (!rx_rule)
+		return;
+
+	if (!MLX5_CAP_ESW_FLOWTABLE(dmn->mdev, fdb_ipv4_ttl_modify)) {
+		/* Ignore the modify TTL action.
+		 * It is always kept as last HW action.
+		 */
+		attr->modify_actions--;
+		return;
+	}
+
+	if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB)
+		/* Due to a HW bug on some devices, modifying TTL on RX flows
+		 * will cause an incorrect checksum calculation. In such cases
+		 * we will use a FW table to recalculate the checksum.
+		 */
+		*recalc_cs_required = true;
+}
+
 static void dr_action_print_sequence(struct mlx5dr_domain *dmn,
 				     struct mlx5dr_action *actions[],
 				     int last_idx)
@@ -650,8 +681,9 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 		case DR_ACTION_TYP_MODIFY_HDR:
 			attr.modify_index = action->rewrite->index;
 			attr.modify_actions = action->rewrite->num_of_actions;
-			recalc_cs_required = action->rewrite->modify_ttl &&
-					     !mlx5dr_ste_supp_ttl_cs_recalc(&dmn->info.caps);
+			if (action->rewrite->modify_ttl)
+				dr_action_modify_ttl_adjust(dmn, &attr, rx_rule,
+							    &recalc_cs_required);
 			break;
 		case DR_ACTION_TYP_L2_TO_TNL_L2:
 		case DR_ACTION_TYP_L2_TO_TNL_L3:
@@ -732,12 +764,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 	*new_hw_ste_arr_sz = nic_matcher->num_of_builders;
 	last_ste = ste_arr + DR_STE_SIZE * (nic_matcher->num_of_builders - 1);
 
-	/* Due to a HW bug in some devices, modifying TTL on RX flows will
-	 * cause an incorrect checksum calculation. In this case we will
-	 * use a FW table to recalculate.
-	 */
-	if (dmn->type == MLX5DR_DOMAIN_TYPE_FDB &&
-	    rx_rule && recalc_cs_required && dest_action) {
+	if (recalc_cs_required && dest_action) {
 		ret = dr_action_handle_cs_recalc(dmn, dest_action, &attr.final_icm_addr);
 		if (ret) {
 			mlx5dr_err(dmn,
@@ -842,7 +869,8 @@ struct mlx5dr_action *
 mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				   struct mlx5dr_action_dest *dests,
 				   u32 num_of_dests,
-				   bool ignore_flow_level)
+				   bool ignore_flow_level,
+				   u32 flow_source)
 {
 	struct mlx5dr_cmd_flow_destination_hw_info *hw_dests;
 	struct mlx5dr_action **ref_actions;
@@ -914,7 +942,8 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				      reformat_req,
 				      &action->dest_tbl->fw_tbl.id,
 				      &action->dest_tbl->fw_tbl.group_id,
-				      ignore_flow_level);
+				      ignore_flow_level,
+				      flow_source);
 	if (ret)
 		goto free_action;
 
@@ -1556,12 +1585,6 @@ dr_action_modify_check_is_ttl_modify(const void *sw_action)
 	return sw_field == MLX5_ACTION_IN_FIELD_OUT_IP_TTL;
 }
 
-static bool dr_action_modify_ttl_ignore(struct mlx5dr_domain *dmn)
-{
-	return !mlx5dr_ste_supp_ttl_cs_recalc(&dmn->info.caps) &&
-	       !MLX5_CAP_ESW_FLOWTABLE(dmn->mdev, fdb_ipv4_ttl_modify);
-}
-
 static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 					    u32 max_hw_actions,
 					    u32 num_sw_actions,
@@ -1573,6 +1596,7 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 	const struct mlx5dr_ste_action_modify_field *hw_dst_action_info;
 	const struct mlx5dr_ste_action_modify_field *hw_src_action_info;
 	struct mlx5dr_domain *dmn = action->rewrite->dmn;
+	__be64 *modify_ttl_sw_action = NULL;
 	int ret, i, hw_idx = 0;
 	__be64 *sw_action;
 	__be64 hw_action;
@@ -1585,8 +1609,14 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 	action->rewrite->allow_rx = 1;
 	action->rewrite->allow_tx = 1;
 
-	for (i = 0; i < num_sw_actions; i++) {
-		sw_action = &sw_actions[i];
+	for (i = 0; i < num_sw_actions || modify_ttl_sw_action; i++) {
+		/* modify TTL is handled separately, as a last action */
+		if (i == num_sw_actions) {
+			sw_action = modify_ttl_sw_action;
+			modify_ttl_sw_action = NULL;
+		} else {
+			sw_action = &sw_actions[i];
+		}
 
 		ret = dr_action_modify_check_field_limitation(action,
 							      sw_action);
@@ -1595,10 +1625,9 @@ static int dr_actions_convert_modify_header(struct mlx5dr_action *action,
 
 		if (!(*modify_ttl) &&
 		    dr_action_modify_check_is_ttl_modify(sw_action)) {
-			if (dr_action_modify_ttl_ignore(dmn))
-				continue;
-
+			modify_ttl_sw_action = sw_action;
 			*modify_ttl = true;
+			continue;
 		}
 
 		/* Convert SW action to HW action */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
index 68a4c32d5f34..f05ef0cd54ba 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c
@@ -104,7 +104,8 @@ int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,
 			    bool reformat_req,
 			    u32 *tbl_id,
 			    u32 *group_id,
-			    bool ignore_flow_level)
+			    bool ignore_flow_level,
+			    u32 flow_source)
 {
 	struct mlx5dr_cmd_create_flow_table_attr ft_attr = {};
 	struct mlx5dr_cmd_fte_info fte_info = {};
@@ -139,6 +140,7 @@ int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,
 	fte_info.val = val;
 	fte_info.dest_arr = dest;
 	fte_info.ignore_flow_level = ignore_flow_level;
+	fte_info.flow_context.flow_source = flow_source;
 
 	ret = mlx5dr_cmd_set_fte(dmn->mdev, 0, 0, &ft_info, *group_id, &fte_info);
 	if (ret) {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
index 5a322335f204..2010d4ac6519 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
@@ -420,7 +420,7 @@ dr_ste_v0_set_actions_tx(struct mlx5dr_domain *dmn,
 	 * encapsulation. The reason for that is that we support
 	 * modify headers for outer headers only
 	 */
-	if (action_type_set[DR_ACTION_TYP_MODIFY_HDR]) {
+	if (action_type_set[DR_ACTION_TYP_MODIFY_HDR] && attr->modify_actions) {
 		dr_ste_v0_set_entry_type(last_ste, DR_STE_TYPE_MODIFY_PKT);
 		dr_ste_v0_set_rewrite_actions(last_ste,
 					      attr->modify_actions,
@@ -513,7 +513,7 @@ dr_ste_v0_set_actions_rx(struct mlx5dr_domain *dmn,
 		}
 	}
 
-	if (action_type_set[DR_ACTION_TYP_MODIFY_HDR]) {
+	if (action_type_set[DR_ACTION_TYP_MODIFY_HDR] && attr->modify_actions) {
 		if (dr_ste_v0_get_entry_type(last_ste) == DR_STE_TYPE_MODIFY_PKT)
 			dr_ste_v0_arr_init_next(&last_ste,
 						added_stes,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 46866a5fc5ca..98320e3945ad 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1461,7 +1461,8 @@ int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,
 			    bool reformat_req,
 			    u32 *tbl_id,
 			    u32 *group_id,
-			    bool ignore_flow_level);
+			    bool ignore_flow_level,
+			    u32 flow_source);
 void mlx5dr_fw_destroy_md_tbl(struct mlx5dr_domain *dmn, u32 tbl_id,
 			      u32 group_id);
 #endif  /* _DR_TYPES_H_ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index 045b0cf90063..728f81882589 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -520,6 +520,7 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 	} else if (num_term_actions > 1) {
 		bool ignore_flow_level =
 			!!(fte->action.flags & FLOW_ACT_IGNORE_FLOW_LEVEL);
+		u32 flow_source = fte->flow_context.flow_source;
 
 		if (num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX ||
 		    fs_dr_num_actions == MLX5_FLOW_CONTEXT_ACTION_MAX) {
@@ -529,7 +530,8 @@ static int mlx5_cmd_dr_create_fte(struct mlx5_flow_root_namespace *ns,
 		tmp_action = mlx5dr_action_create_mult_dest_tbl(domain,
 								term_actions,
 								num_term_actions,
-								ignore_flow_level);
+								ignore_flow_level,
+								flow_source);
 		if (!tmp_action) {
 			err = -EOPNOTSUPP;
 			goto free_actions;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
index ec5cbec0d455..7626c85643b1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/mlx5dr.h
@@ -99,7 +99,8 @@ struct mlx5dr_action *
 mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				   struct mlx5dr_action_dest *dests,
 				   u32 num_of_dests,
-				   bool ignore_flow_level);
+				   bool ignore_flow_level,
+				   u32 flow_source);
 
 struct mlx5dr_action *mlx5dr_action_create_drop(void);