Merge branch 'ct-offload' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

drivers/infiniband/hw/mlx5/main.c

@@ -3570,7 +3570,8 @@ static void mlx5_ib_set_rule_source_port(struct mlx5_ib_dev *dev,
 		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
 				    misc_parameters_2);
 
-		MLX5_SET_TO_ONES(fte_match_set_misc2, misc, metadata_reg_c_0);
+		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
+			 mlx5_eswitch_get_vport_metadata_mask());
 	} else {
 		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
 				    misc_parameters);

drivers/net/ethernet/mellanox/mlx5/core/Makefile

@@ -34,7 +34,7 @@ mlx5_core-$(CONFIG_MLX5_EN_ARFS)     += en_arfs.o
 mlx5_core-$(CONFIG_MLX5_EN_RXNFC)    += en_fs_ethtool.o
 mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o en/port_buffer.o
 mlx5_core-$(CONFIG_MLX5_ESWITCH)     += en_rep.o en_tc.o en/tc_tun.o lib/port_tun.o lag_mp.o \
-					lib/geneve.o en/tc_tun_vxlan.o en/tc_tun_gre.o \
+					lib/geneve.o en/mapping.o en/tc_tun_vxlan.o en/tc_tun_gre.o \
 					en/tc_tun_geneve.o diag/en_tc_tracepoint.o
 mlx5_core-$(CONFIG_PCI_HYPERV_INTERFACE) += en/hv_vhca_stats.o
 

drivers/net/ethernet/mellanox/mlx5/core/en/mapping.c

+// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB
+/* Copyright (c) 2018 Mellanox Technologies */
+
+#include <linux/jhash.h>
+#include <linux/slab.h>
+#include <linux/xarray.h>
+#include <linux/hashtable.h>
+
+#include "mapping.h"
+
+#define MAPPING_GRACE_PERIOD 2000
+
+struct mapping_ctx {
+	struct xarray xarray;
+	DECLARE_HASHTABLE(ht, 8);
+	struct mutex lock; /* Guards hashtable and xarray */
+	unsigned long max_id;
+	size_t data_size;
+	bool delayed_removal;
+	struct delayed_work dwork;
+	struct list_head pending_list;
+	spinlock_t pending_list_lock; /* Guards pending list */
+};
+
+struct mapping_item {
+	struct rcu_head rcu;
+	struct list_head list;
+	unsigned long timeout;
+	struct hlist_node node;
+	int cnt;
+	u32 id;
+	char data[];
+};
+
+int mapping_add(struct mapping_ctx *ctx, void *data, u32 *id)
+{
+	struct mapping_item *mi;
+	int err = -ENOMEM;
+	u32 hash_key;
+
+	mutex_lock(&ctx->lock);
+
+	hash_key = jhash(data, ctx->data_size, 0);
+	hash_for_each_possible(ctx->ht, mi, node, hash_key) {
+		if (!memcmp(data, mi->data, ctx->data_size))
+			goto attach;
+	}
+
+	mi = kzalloc(sizeof(*mi) + ctx->data_size, GFP_KERNEL);
+	if (!mi)
+		goto err_alloc;
+
+	memcpy(mi->data, data, ctx->data_size);
+	hash_add(ctx->ht, &mi->node, hash_key);
+
+	err = xa_alloc(&ctx->xarray, &mi->id, mi, XA_LIMIT(1, ctx->max_id),
+		       GFP_KERNEL);
+	if (err)
+		goto err_assign;
+attach:
+	++mi->cnt;
+	*id = mi->id;
+
+	mutex_unlock(&ctx->lock);
+
+	return 0;
+
+err_assign:
+	hash_del(&mi->node);
+	kfree(mi);
+err_alloc:
+	mutex_unlock(&ctx->lock);
+
+	return err;
+}
+
+static void mapping_remove_and_free(struct mapping_ctx *ctx,
+				    struct mapping_item *mi)
+{
+	xa_erase(&ctx->xarray, mi->id);
+	kfree_rcu(mi, rcu);
+}
+
+static void mapping_free_item(struct mapping_ctx *ctx,
+			      struct mapping_item *mi)
+{
+	if (!ctx->delayed_removal) {
+		mapping_remove_and_free(ctx, mi);
+		return;
+	}
+
+	mi->timeout = jiffies + msecs_to_jiffies(MAPPING_GRACE_PERIOD);
+
+	spin_lock(&ctx->pending_list_lock);
+	list_add_tail(&mi->list, &ctx->pending_list);
+	spin_unlock(&ctx->pending_list_lock);
+
+	schedule_delayed_work(&ctx->dwork, MAPPING_GRACE_PERIOD);
+}
+
+int mapping_remove(struct mapping_ctx *ctx, u32 id)
+{
+	unsigned long index = id;
+	struct mapping_item *mi;
+	int err = -ENOENT;
+
+	mutex_lock(&ctx->lock);
+	mi = xa_load(&ctx->xarray, index);
+	if (!mi)
+		goto out;
+	err = 0;
+
+	if (--mi->cnt > 0)
+		goto out;
+
+	hash_del(&mi->node);
+	mapping_free_item(ctx, mi);
+out:
+	mutex_unlock(&ctx->lock);
+
+	return err;
+}
+
+int mapping_find(struct mapping_ctx *ctx, u32 id, void *data)
+{
+	unsigned long index = id;
+	struct mapping_item *mi;
+	int err = -ENOENT;
+
+	rcu_read_lock();
+	mi = xa_load(&ctx->xarray, index);
+	if (!mi)
+		goto err_find;
+
+	memcpy(data, mi->data, ctx->data_size);
+	err = 0;
+
+err_find:
+	rcu_read_unlock();
+	return err;
+}
+
+static void
+mapping_remove_and_free_list(struct mapping_ctx *ctx, struct list_head *list)
+{
+	struct mapping_item *mi;
+
+	list_for_each_entry(mi, list, list)
+		mapping_remove_and_free(ctx, mi);
+}
+
+static void mapping_work_handler(struct work_struct *work)
+{
+	unsigned long min_timeout = 0, now = jiffies;
+	struct mapping_item *mi, *next;
+	LIST_HEAD(pending_items);
+	struct mapping_ctx *ctx;
+
+	ctx = container_of(work, struct mapping_ctx, dwork.work);
+
+	spin_lock(&ctx->pending_list_lock);
+	list_for_each_entry_safe(mi, next, &ctx->pending_list, list) {
+		if (time_after(now, mi->timeout))
+			list_move(&mi->list, &pending_items);
+		else if (!min_timeout ||
+			 time_before(mi->timeout, min_timeout))
+			min_timeout = mi->timeout;
+	}
+	spin_unlock(&ctx->pending_list_lock);
+
+	mapping_remove_and_free_list(ctx, &pending_items);
+
+	if (min_timeout)
+		schedule_delayed_work(&ctx->dwork, abs(min_timeout - now));
+}
+
+static void mapping_flush_work(struct mapping_ctx *ctx)
+{
+	if (!ctx->delayed_removal)
+		return;
+
+	cancel_delayed_work_sync(&ctx->dwork);
+	mapping_remove_and_free_list(ctx, &ctx->pending_list);
+}
+
+struct mapping_ctx *
+mapping_create(size_t data_size, u32 max_id, bool delayed_removal)
+{
+	struct mapping_ctx *ctx;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return ERR_PTR(-ENOMEM);
+
+	ctx->max_id = max_id ? max_id : UINT_MAX;
+	ctx->data_size = data_size;
+
+	if (delayed_removal) {
+		INIT_DELAYED_WORK(&ctx->dwork, mapping_work_handler);
+		INIT_LIST_HEAD(&ctx->pending_list);
+		spin_lock_init(&ctx->pending_list_lock);
+		ctx->delayed_removal = true;
+	}
+
+	mutex_init(&ctx->lock);
+	xa_init_flags(&ctx->xarray, XA_FLAGS_ALLOC1);
+
+	return ctx;
+}
+
+void mapping_destroy(struct mapping_ctx *ctx)
+{
+	mapping_flush_work(ctx);
+	xa_destroy(&ctx->xarray);
+	mutex_destroy(&ctx->lock);
+
+	kfree(ctx);
+}

drivers/net/ethernet/mellanox/mlx5/core/en/mapping.h

+/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
+/* Copyright (c) 2019 Mellanox Technologies */
+
+#ifndef __MLX5_MAPPING_H__
+#define __MLX5_MAPPING_H__
+
+struct mapping_ctx;
+
+int mapping_add(struct mapping_ctx *ctx, void *data, u32 *id);
+int mapping_remove(struct mapping_ctx *ctx, u32 id);
+int mapping_find(struct mapping_ctx *ctx, u32 id, void *data);
+
+/* mapping uses an xarray to map data to ids in add(), and for find().
+ * For locking, it uses a internal xarray spin lock for add()/remove(),
+ * find() uses rcu_read_lock().
+ * Choosing delayed_removal postpones the removal of a previously mapped
+ * id by MAPPING_GRACE_PERIOD milliseconds.
+ * This is to avoid races against hardware, where we mark the packet in
+ * hardware with a previous id, and quick remove() and add() reusing the same
+ * previous id. Then find() will get the new mapping instead of the old
+ * which was used to mark the packet.
+ */
+struct mapping_ctx *mapping_create(size_t data_size, u32 max_id,
+				   bool delayed_removal);
+void mapping_destroy(struct mapping_ctx *ctx);
+
+#endif /* __MLX5_MAPPING_H__ */

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c

@@ -469,10 +469,15 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev,
 		       struct mlx5e_priv *priv,
 		       struct mlx5_flow_spec *spec,
 		       struct flow_cls_offload *f,
-		       void *headers_c,
-		       void *headers_v, u8 *match_level)
+		       u8 *match_level)
 {
 	struct mlx5e_tc_tunnel *tunnel = mlx5e_get_tc_tun(filter_dev);
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+				       outer_headers);
+	void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
+				       outer_headers);
+	struct netlink_ext_ack *extack = f->common.extack;
 	int err = 0;
 
 	if (!tunnel) {
@@ -499,6 +504,109 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev,
 			goto out;
 	}
 
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) {
+		struct flow_match_control match;
+		u16 addr_type;
+
+		flow_rule_match_enc_control(rule, &match);
+		addr_type = match.key->addr_type;
+
+		/* For tunnel addr_type used same key id`s as for non-tunnel */
+		if (addr_type == FLOW_DISSECTOR_KEY_IPV4_ADDRS) {
+			struct flow_match_ipv4_addrs match;
+
+			flow_rule_match_enc_ipv4_addrs(rule, &match);
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 src_ipv4_src_ipv6.ipv4_layout.ipv4,
+				 ntohl(match.mask->src));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 src_ipv4_src_ipv6.ipv4_layout.ipv4,
+				 ntohl(match.key->src));
+
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+				 dst_ipv4_dst_ipv6.ipv4_layout.ipv4,
+				 ntohl(match.mask->dst));
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+				 dst_ipv4_dst_ipv6.ipv4_layout.ipv4,
+				 ntohl(match.key->dst));
+
+			MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c,
+					 ethertype);
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype,
+				 ETH_P_IP);
+		} else if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+			struct flow_match_ipv6_addrs match;
+
+			flow_rule_match_enc_ipv6_addrs(rule, &match);
+			memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+					    src_ipv4_src_ipv6.ipv6_layout.ipv6),
+			       &match.mask->src, MLX5_FLD_SZ_BYTES(ipv6_layout,
+								   ipv6));
+			memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+					    src_ipv4_src_ipv6.ipv6_layout.ipv6),
+			       &match.key->src, MLX5_FLD_SZ_BYTES(ipv6_layout,
+								  ipv6));
+
+			memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_c,
+					    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
+			       &match.mask->dst, MLX5_FLD_SZ_BYTES(ipv6_layout,
+								   ipv6));
+			memcpy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+					    dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
+			       &match.key->dst, MLX5_FLD_SZ_BYTES(ipv6_layout,
+								  ipv6));
+
+			MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c,
+					 ethertype);
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype,
+				 ETH_P_IPV6);
+		}
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) {
+		struct flow_match_ip match;
+
+		flow_rule_match_enc_ip(rule, &match);
+		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_ecn,
+			 match.mask->tos & 0x3);
+		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ecn,
+			 match.key->tos & 0x3);
+
+		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_dscp,
+			 match.mask->tos >> 2);
+		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_dscp,
+			 match.key->tos  >> 2);
+
+		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ttl_hoplimit,
+			 match.mask->ttl);
+		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ttl_hoplimit,
+			 match.key->ttl);
+
+		if (match.mask->ttl &&
+		    !MLX5_CAP_ESW_FLOWTABLE_FDB
+			(priv->mdev,
+			 ft_field_support.outer_ipv4_ttl)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Matching on TTL is not supported");
+			err = -EOPNOTSUPP;
+			goto out;
+		}
+	}
+
+	/* Enforce DMAC when offloading incoming tunneled flows.
+	 * Flow counters require a match on the DMAC.
+	 */
+	MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_47_16);
+	MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, dmac_15_0);
+	ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, headers_v,
+				     dmac_47_16), priv->netdev->dev_addr);
+
+	/* let software handle IP fragments */
+	MLX5_SET(fte_match_set_lyr_2_4, headers_c, frag, 1);
+	MLX5_SET(fte_match_set_lyr_2_4, headers_v, frag, 0);
+
+	return 0;
+
 out:
 	return err;
 }

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h

@@ -76,8 +76,7 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev,
 		       struct mlx5e_priv *priv,
 		       struct mlx5_flow_spec *spec,
 		       struct flow_cls_offload *f,
-		       void *headers_c,
-		       void *headers_v, u8 *match_level);
+		       u8 *match_level);
 
 int mlx5e_tc_tun_parse_udp_ports(struct mlx5e_priv *priv,
 				 struct mlx5_flow_spec *spec,

drivers/net/ethernet/mellanox/mlx5/core/en_rep.c

@@ -1952,7 +1952,7 @@ static const struct mlx5e_profile mlx5e_rep_profile = {
 	.update_rx		= mlx5e_update_rep_rx,
 	.update_stats           = mlx5e_update_ndo_stats,
 	.rx_handlers.handle_rx_cqe       = mlx5e_handle_rx_cqe_rep,
-	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq,
+	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq_rep,
 	.max_tc			= 1,
 	.rq_groups		= MLX5E_NUM_RQ_GROUPS(REGULAR),
 	.stats_grps		= mlx5e_rep_stats_grps,
@@ -1972,7 +1972,7 @@ static const struct mlx5e_profile mlx5e_uplink_rep_profile = {
 	.update_stats           = mlx5e_update_ndo_stats,
 	.update_carrier	        = mlx5e_update_carrier,
 	.rx_handlers.handle_rx_cqe       = mlx5e_handle_rx_cqe_rep,
-	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq,
+	.rx_handlers.handle_rx_cqe_mpwqe = mlx5e_handle_rx_cqe_mpwrq_rep,
 	.max_tc			= MLX5E_MAX_NUM_TC,
 	.rq_groups		= MLX5E_NUM_RQ_GROUPS(REGULAR),
 	.stats_grps		= mlx5e_ul_rep_stats_grps,

drivers/net/ethernet/mellanox/mlx5/core/en_rep.h

@@ -81,6 +81,11 @@ struct mlx5_rep_uplink_priv {
 	struct mutex                unready_flows_lock;
 	struct list_head            unready_flows;
 	struct work_struct          reoffload_flows_work;
+
+	/* maps tun_info to a unique id*/
+	struct mapping_ctx *tunnel_mapping;
+	/* maps tun_enc_opts to a unique id*/
+	struct mapping_ctx *tunnel_enc_opts_mapping;
 };
 
 struct mlx5e_rep_priv {
@@ -192,6 +197,8 @@ int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv);
 void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv);
 
 void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe);
+void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq,
+				   struct mlx5_cqe64 *cqe);
 
 int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv,
 				 struct mlx5e_encap_entry *e);

drivers/net/ethernet/mellanox/mlx5/core/en_rx.c

@@ -1195,6 +1195,7 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 	struct mlx5e_priv *priv = netdev_priv(netdev);
 	struct mlx5e_rep_priv *rpriv  = priv->ppriv;
 	struct mlx5_eswitch_rep *rep = rpriv->rep;
+	struct mlx5e_tc_update_priv tc_priv = {};
 	struct mlx5_wq_cyc *wq = &rq->wqe.wq;
 	struct mlx5e_wqe_frag_info *wi;
 	struct sk_buff *skb;
@@ -1227,13 +1228,78 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 	if (rep->vlan && skb_vlan_tag_present(skb))
 		skb_vlan_pop(skb);
 
+	if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv))
+		goto free_wqe;
+
 	napi_gro_receive(rq->cq.napi, skb);
 
+	mlx5_tc_rep_post_napi_receive(&tc_priv);
+
 free_wqe:
 	mlx5e_free_rx_wqe(rq, wi, true);
 wq_cyc_pop:
 	mlx5_wq_cyc_pop(wq);
 }
+
+void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq,
+				   struct mlx5_cqe64 *cqe)
+{
+	u16 cstrides       = mpwrq_get_cqe_consumed_strides(cqe);
+	u16 wqe_id         = be16_to_cpu(cqe->wqe_id);
+	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[wqe_id];
+	u16 stride_ix      = mpwrq_get_cqe_stride_index(cqe);
+	u32 wqe_offset     = stride_ix << rq->mpwqe.log_stride_sz;
+	u32 head_offset    = wqe_offset & (PAGE_SIZE - 1);
+	u32 page_idx       = wqe_offset >> PAGE_SHIFT;
+	struct mlx5e_tc_update_priv tc_priv = {};
+	struct mlx5e_rx_wqe_ll *wqe;
+	struct mlx5_wq_ll *wq;
+	struct sk_buff *skb;
+	u16 cqe_bcnt;
+
+	wi->consumed_strides += cstrides;
+
+	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {
+		trigger_report(rq, cqe);
+		rq->stats->wqe_err++;
+		goto mpwrq_cqe_out;
+	}
+
+	if (unlikely(mpwrq_is_filler_cqe(cqe))) {
+		struct mlx5e_rq_stats *stats = rq->stats;
+
+		stats->mpwqe_filler_cqes++;
+		stats->mpwqe_filler_strides += cstrides;
+		goto mpwrq_cqe_out;
+	}
+
+	cqe_bcnt = mpwrq_get_cqe_byte_cnt(cqe);
+
+	skb = INDIRECT_CALL_2(rq->mpwqe.skb_from_cqe_mpwrq,
+			      mlx5e_skb_from_cqe_mpwrq_linear,
+			      mlx5e_skb_from_cqe_mpwrq_nonlinear,
+			      rq, wi, cqe_bcnt, head_offset, page_idx);
+	if (!skb)
+		goto mpwrq_cqe_out;
+
+	mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb);
+
+	if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv))
+		goto mpwrq_cqe_out;
+
+	napi_gro_receive(rq->cq.napi, skb);
+
+	mlx5_tc_rep_post_napi_receive(&tc_priv);
+
+mpwrq_cqe_out:
+	if (likely(wi->consumed_strides < rq->mpwqe.num_strides))
+		return;
+
+	wq  = &rq->mpwqe.wq;
+	wqe = mlx5_wq_ll_get_wqe(wq, wqe_id);
+	mlx5e_free_rx_mpwqe(rq, wi, true);
+	mlx5_wq_ll_pop(wq, cqe->wqe_id, &wqe->next.next_wqe_index);
+}
 #endif
 
 struct sk_buff *

drivers/net/ethernet/mellanox/mlx5/core/en_tc.h

@@ -91,9 +91,54 @@ int mlx5e_tc_num_filters(struct mlx5e_priv *priv, unsigned long flags);
 
 void mlx5e_tc_reoffload_flows_work(struct work_struct *work);
 
+enum mlx5e_tc_attr_to_reg {
+	CHAIN_TO_REG,
+	TUNNEL_TO_REG,
+};
+
+struct mlx5e_tc_attr_to_reg_mapping {
+	int mfield; /* rewrite field */
+	int moffset; /* offset of mfield */
+	int mlen; /* bytes to rewrite/match */
+
+	int soffset; /* offset of spec for match */
+};
+
+extern struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[];
+
 bool mlx5e_is_valid_eswitch_fwd_dev(struct mlx5e_priv *priv,
 				    struct net_device *out_dev);
 
+struct mlx5e_tc_update_priv {
+	struct net_device *tun_dev;
+};
+
+bool mlx5e_tc_rep_update_skb(struct mlx5_cqe64 *cqe, struct sk_buff *skb,
+			     struct mlx5e_tc_update_priv *tc_priv);
+
+void mlx5_tc_rep_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv);
+
+struct mlx5e_tc_mod_hdr_acts {
+	int num_actions;
+	int max_actions;
+	void *actions;
+};
+
+int mlx5e_tc_match_to_reg_set(struct mlx5_core_dev *mdev,
+			      struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts,
+			      enum mlx5e_tc_attr_to_reg type,
+			      u32 data);
+
+void mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec,
+				 enum mlx5e_tc_attr_to_reg type,
+				 u32 data,
+				 u32 mask);
+
+int alloc_mod_hdr_actions(struct mlx5_core_dev *mdev,
+			  int namespace,
+			  struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts);
+void dealloc_mod_hdr_actions(struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts);
+
 #else /* CONFIG_MLX5_ESWITCH */
 static inline int  mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; }
 static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {}

drivers/net/ethernet/mellanox/mlx5/core/eswitch.h

@@ -197,6 +197,10 @@ struct mlx5_eswitch_fdb {
 };
 
 struct mlx5_esw_offload {
+	struct mlx5_flow_table *ft_offloads_restore;
+	struct mlx5_flow_group *restore_group;
+	struct mlx5_modify_hdr *restore_copy_hdr_id;
+
 	struct mlx5_flow_table *ft_offloads;
 	struct mlx5_flow_group *vport_rx_group;
 	struct mlx5_eswitch_rep *vport_reps;
@@ -636,6 +640,11 @@ esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw,
 int mlx5_esw_vport_tbl_get(struct mlx5_eswitch *esw);
 void mlx5_esw_vport_tbl_put(struct mlx5_eswitch *esw);
 
+struct mlx5_flow_handle *
+esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag);
+u32
+esw_get_max_restore_tag(struct mlx5_eswitch *esw);
+
 #else  /* CONFIG_MLX5_ESWITCH */
 /* eswitch API stubs */
 static inline int  mlx5_eswitch_init(struct mlx5_core_dev *dev) { return 0; }
@@ -651,6 +660,12 @@ static inline const u32 *mlx5_esw_query_functions(struct mlx5_core_dev *dev)
 
 static inline void mlx5_eswitch_update_num_of_vfs(struct mlx5_eswitch *esw, const int num_vfs) {}
 
+static struct mlx5_flow_handle *
+esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif /* __MLX5_ESWITCH_H__ */

drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.c

@@ -6,14 +6,17 @@
 #include <linux/mlx5/fs.h>
 
 #include "eswitch_offloads_chains.h"
+#include "en/mapping.h"
 #include "mlx5_core.h"
 #include "fs_core.h"
 #include "eswitch.h"
 #include "en.h"
+#include "en_tc.h"
 
 #define esw_chains_priv(esw) ((esw)->fdb_table.offloads.esw_chains_priv)
 #define esw_chains_lock(esw) (esw_chains_priv(esw)->lock)
 #define esw_chains_ht(esw) (esw_chains_priv(esw)->chains_ht)
+#define esw_chains_mapping(esw) (esw_chains_priv(esw)->chains_mapping)
 #define esw_prios_ht(esw) (esw_chains_priv(esw)->prios_ht)
 #define fdb_pool_left(esw) (esw_chains_priv(esw)->fdb_left)
 #define tc_slow_fdb(esw) ((esw)->fdb_table.offloads.slow_fdb)
@@ -43,6 +46,7 @@ struct mlx5_esw_chains_priv {
 	struct mutex lock;
 
 	struct mlx5_flow_table *tc_end_fdb;
+	struct mapping_ctx *chains_mapping;
 
 	int fdb_left[ARRAY_SIZE(ESW_POOLS)];
 };
@@ -53,9 +57,12 @@ struct fdb_chain {
 	u32 chain;
 
 	int ref;
+	int id;
 
 	struct mlx5_eswitch *esw;
 	struct list_head prios_list;
+	struct mlx5_flow_handle *restore_rule;
+	struct mlx5_modify_hdr *miss_modify_hdr;
 };
 
 struct fdb_prio_key {
@@ -261,6 +268,70 @@ mlx5_esw_chains_destroy_fdb_table(struct mlx5_eswitch *esw,
 	mlx5_destroy_flow_table(fdb);
 }
 
+static int
+create_fdb_chain_restore(struct fdb_chain *fdb_chain)
+{
+	char modact[MLX5_UN_SZ_BYTES(set_action_in_add_action_in_auto)];
+	struct mlx5_eswitch *esw = fdb_chain->esw;
+	struct mlx5_modify_hdr *mod_hdr;
+	u32 index;
+	int err;
+
+	if (fdb_chain->chain == mlx5_esw_chains_get_ft_chain(esw))
+		return 0;
+
+	err = mapping_add(esw_chains_mapping(esw), &fdb_chain->chain, &index);
+	if (err)
+		return err;
+	if (index == MLX5_FS_DEFAULT_FLOW_TAG) {
+		/* we got the special default flow tag id, so we won't know
+		 * if we actually marked the packet with the restore rule
+		 * we create.
+		 *
+		 * This case isn't possible with MLX5_FS_DEFAULT_FLOW_TAG = 0.
+		 */
+		err = mapping_add(esw_chains_mapping(esw),
+				  &fdb_chain->chain, &index);
+		mapping_remove(esw_chains_mapping(esw),
+			       MLX5_FS_DEFAULT_FLOW_TAG);
+		if (err)
+			return err;
+	}
+
+	fdb_chain->id = index;
+
+	MLX5_SET(set_action_in, modact, action_type, MLX5_ACTION_TYPE_SET);
+	MLX5_SET(set_action_in, modact, field,
+		 mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mfield);
+	MLX5_SET(set_action_in, modact, offset,
+		 mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].moffset * 8);
+	MLX5_SET(set_action_in, modact, length,
+		 mlx5e_tc_attr_to_reg_mappings[CHAIN_TO_REG].mlen * 8);
+	MLX5_SET(set_action_in, modact, data, fdb_chain->id);
+	mod_hdr = mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_FDB,
+					   1, modact);
+	if (IS_ERR(mod_hdr)) {
+		err = PTR_ERR(mod_hdr);
+		goto err_mod_hdr;
+	}
+	fdb_chain->miss_modify_hdr = mod_hdr;
+
+	fdb_chain->restore_rule = esw_add_restore_rule(esw, fdb_chain->id);
+	if (IS_ERR(fdb_chain->restore_rule)) {
+		err = PTR_ERR(fdb_chain->restore_rule);
+		goto err_rule;
+	}
+
+	return 0;
+
+err_rule:
+	mlx5_modify_header_dealloc(esw->dev, fdb_chain->miss_modify_hdr);
+err_mod_hdr:
+	/* Datapath can't find this mapping, so we can safely remove it */
+	mapping_remove(esw_chains_mapping(esw), fdb_chain->id);
+	return err;
+}
+
 static struct fdb_chain *
 mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 {
@@ -275,6 +346,10 @@ mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 	fdb_chain->chain = chain;
 	INIT_LIST_HEAD(&fdb_chain->prios_list);
 
+	err = create_fdb_chain_restore(fdb_chain);
+	if (err)
+		goto err_restore;
+
 	err = rhashtable_insert_fast(&esw_chains_ht(esw), &fdb_chain->node,
 				     chain_params);
 	if (err)
@@ -283,6 +358,12 @@ mlx5_esw_chains_create_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 	return fdb_chain;
 
 err_insert:
+	if (fdb_chain->chain != mlx5_esw_chains_get_ft_chain(esw)) {
+		mlx5_del_flow_rules(fdb_chain->restore_rule);
+		mlx5_modify_header_dealloc(esw->dev,
+					   fdb_chain->miss_modify_hdr);
+	}
+err_restore:
 	kvfree(fdb_chain);
 	return ERR_PTR(err);
 }
@@ -294,6 +375,15 @@ mlx5_esw_chains_destroy_fdb_chain(struct fdb_chain *fdb_chain)
 
 	rhashtable_remove_fast(&esw_chains_ht(esw), &fdb_chain->node,
 			       chain_params);
+
+	if (fdb_chain->chain != mlx5_esw_chains_get_ft_chain(esw)) {
+		mlx5_del_flow_rules(fdb_chain->restore_rule);
+		mlx5_modify_header_dealloc(esw->dev,
+					   fdb_chain->miss_modify_hdr);
+
+		mapping_remove(esw_chains_mapping(esw), fdb_chain->id);
+	}
+
 	kvfree(fdb_chain);
 }
 
@@ -316,10 +406,12 @@ mlx5_esw_chains_get_fdb_chain(struct mlx5_eswitch *esw, u32 chain)
 }
 
 static struct mlx5_flow_handle *
-mlx5_esw_chains_add_miss_rule(struct mlx5_flow_table *fdb,
+mlx5_esw_chains_add_miss_rule(struct fdb_chain *fdb_chain,
+			      struct mlx5_flow_table *fdb,
 			      struct mlx5_flow_table *next_fdb)
 {
 	static const struct mlx5_flow_spec spec = {};
+	struct mlx5_eswitch *esw = fdb_chain->esw;
 	struct mlx5_flow_destination dest = {};
 	struct mlx5_flow_act act = {};
 
@@ -328,6 +420,11 @@ mlx5_esw_chains_add_miss_rule(struct mlx5_flow_table *fdb,
 	dest.type  = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 	dest.ft = next_fdb;
 
+	if (fdb_chain->chain != mlx5_esw_chains_get_ft_chain(esw)) {
+		act.modify_hdr = fdb_chain->miss_modify_hdr;
+		act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
+	}
+
 	return mlx5_add_flow_rules(fdb, &spec, &act, &dest, 1);
 }
 
@@ -351,7 +448,8 @@ mlx5_esw_chains_update_prio_prevs(struct fdb_prio *fdb_prio,
 	list_for_each_entry_continue_reverse(pos,
 					     &fdb_chain->prios_list,
 					     list) {
-		miss_rules[n] = mlx5_esw_chains_add_miss_rule(pos->fdb,
+		miss_rules[n] = mlx5_esw_chains_add_miss_rule(fdb_chain,
+							      pos->fdb,
 							      next_fdb);
 		if (IS_ERR(miss_rules[n])) {
 			err = PTR_ERR(miss_rules[n]);
@@ -465,7 +563,7 @@ mlx5_esw_chains_create_fdb_prio(struct mlx5_eswitch *esw,
 	}
 
 	/* Add miss rule to next_fdb */
-	miss_rule = mlx5_esw_chains_add_miss_rule(fdb, next_fdb);
+	miss_rule = mlx5_esw_chains_add_miss_rule(fdb_chain, fdb, next_fdb);
 	if (IS_ERR(miss_rule)) {
 		err = PTR_ERR(miss_rule);
 		goto err_miss_rule;
@@ -630,6 +728,7 @@ mlx5_esw_chains_init(struct mlx5_eswitch *esw)
 	struct mlx5_esw_chains_priv *chains_priv;
 	struct mlx5_core_dev *dev = esw->dev;
 	u32 max_flow_counter, fdb_max;
+	struct mapping_ctx *mapping;
 	int err;
 
 	chains_priv = kzalloc(sizeof(*chains_priv), GFP_KERNEL);
@@ -666,10 +765,20 @@ mlx5_esw_chains_init(struct mlx5_eswitch *esw)
 	if (err)
 		goto init_prios_ht_err;
 
+	mapping = mapping_create(sizeof(u32), esw_get_max_restore_tag(esw),
+				 true);
+	if (IS_ERR(mapping)) {
+		err = PTR_ERR(mapping);
+		goto mapping_err;
+	}
+	esw_chains_mapping(esw) = mapping;
+
 	mutex_init(&esw_chains_lock(esw));
 
 	return 0;
 
+mapping_err:
+	rhashtable_destroy(&esw_prios_ht(esw));
 init_prios_ht_err:
 	rhashtable_destroy(&esw_chains_ht(esw));
 init_chains_ht_err:
@@ -681,6 +790,7 @@ static void
 mlx5_esw_chains_cleanup(struct mlx5_eswitch *esw)
 {
 	mutex_destroy(&esw_chains_lock(esw));
+	mapping_destroy(esw_chains_mapping(esw));
 	rhashtable_destroy(&esw_prios_ht(esw));
 	rhashtable_destroy(&esw_chains_ht(esw));
 
@@ -759,3 +869,17 @@ mlx5_esw_chains_destroy(struct mlx5_eswitch *esw)
 	mlx5_esw_chains_close(esw);
 	mlx5_esw_chains_cleanup(esw);
 }
+
+int mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag,
+				   u32 *chain)
+{
+	int err;
+
+	err = mapping_find(esw_chains_mapping(esw), tag, chain);
+	if (err) {
+		esw_warn(esw->dev, "Can't find chain for tag: %d\n", tag);
+		return -ENOENT;
+	}
+
+	return 0;
+}

drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_chains.h

@@ -28,5 +28,7 @@ mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw);
 int mlx5_esw_chains_create(struct mlx5_eswitch *esw);
 void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw);
 
-#endif /* __ML5_ESW_CHAINS_H__ */
+int
+mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, u32 *chain);
 
+#endif /* __ML5_ESW_CHAINS_H__ */

drivers/net/ethernet/mellanox/mlx5/core/fs_core.c

@@ -111,8 +111,8 @@
 #define ANCHOR_MIN_LEVEL (BY_PASS_MIN_LEVEL + 1)
 
 #define OFFLOADS_MAX_FT 2
-#define OFFLOADS_NUM_PRIOS 1
-#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1)
+#define OFFLOADS_NUM_PRIOS 2
+#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + OFFLOADS_NUM_PRIOS)
 
 #define LAG_PRIO_NUM_LEVELS 1
 #define LAG_NUM_PRIOS 1

include/linux/mlx5/eswitch.h

@@ -71,7 +71,28 @@ enum devlink_eswitch_encap_mode
 mlx5_eswitch_get_encap_mode(const struct mlx5_core_dev *dev);
 
 bool mlx5_eswitch_vport_match_metadata_enabled(const struct mlx5_eswitch *esw);
-u32 mlx5_eswitch_get_vport_metadata_for_match(const struct mlx5_eswitch *esw,
+
+/* Reg C0 usage:
+ * Reg C0 = < ESW_VHCA_ID_BITS(8) | ESW_VPORT BITS(8) | ESW_CHAIN_TAG(16) >
+ *
+ * Highest 8 bits of the reg c0 is the vhca_id, next 8 bits is vport_num,
+ * the rest (lowest 16 bits) is left for tc chain tag restoration.
+ * VHCA_ID + VPORT comprise the SOURCE_PORT matching.
+ */
+#define ESW_VHCA_ID_BITS 8
+#define ESW_VPORT_BITS 8
+#define ESW_SOURCE_PORT_METADATA_BITS (ESW_VHCA_ID_BITS + ESW_VPORT_BITS)
+#define ESW_SOURCE_PORT_METADATA_OFFSET (32 - ESW_SOURCE_PORT_METADATA_BITS)
+#define ESW_CHAIN_TAG_METADATA_BITS (32 - ESW_SOURCE_PORT_METADATA_BITS)
+#define ESW_CHAIN_TAG_METADATA_MASK GENMASK(ESW_CHAIN_TAG_METADATA_BITS - 1,\
+					    0)
+
+static inline u32 mlx5_eswitch_get_vport_metadata_mask(void)
+{
+	return GENMASK(31, 32 - ESW_SOURCE_PORT_METADATA_BITS);
+}
+
+u32 mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw,
 					      u16 vport_num);
 u8 mlx5_eswitch_mode(struct mlx5_eswitch *esw);
 #else  /* CONFIG_MLX5_ESWITCH */
@@ -94,11 +115,17 @@ mlx5_eswitch_vport_match_metadata_enabled(const struct mlx5_eswitch *esw)
 };
 
 static inline u32
-mlx5_eswitch_get_vport_metadata_for_match(const struct mlx5_eswitch *esw,
+mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw,
 					  int vport_num)
 {
 	return 0;
 };
+
+static inline u32
+mlx5_eswitch_get_vport_metadata_mask(void)
+{
+	return 0;
+}
 #endif /* CONFIG_MLX5_ESWITCH */
 
 #endif

include/net/pkt_cls.h

@@ -72,6 +72,10 @@ static inline struct Qdisc *tcf_block_q(struct tcf_block *block)
 
 int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 		 struct tcf_result *res, bool compat_mode);
+int tcf_classify_ingress(struct sk_buff *skb,
+			 const struct tcf_block *ingress_block,
+			 const struct tcf_proto *tp, struct tcf_result *res,
+			 bool compat_mode);
 
 #else
 static inline bool tcf_block_shared(struct tcf_block *block)
@@ -133,6 +137,15 @@ static inline int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 {
 	return TC_ACT_UNSPEC;
 }
+
+static inline int tcf_classify_ingress(struct sk_buff *skb,
+				       const struct tcf_block *ingress_block,
+				       const struct tcf_proto *tp,
+				       struct tcf_result *res, bool compat_mode)
+{
+	return TC_ACT_UNSPEC;
+}
+
 #endif
 
 static inline unsigned long

include/net/sch_generic.h

@@ -1269,6 +1269,7 @@ static inline void psched_ratecfg_getrate(struct tc_ratespec *res,
  */
 struct mini_Qdisc {
 	struct tcf_proto *filter_list;
+	struct tcf_block *block;
 	struct gnet_stats_basic_cpu __percpu *cpu_bstats;
 	struct gnet_stats_queue	__percpu *cpu_qstats;
 	struct rcu_head rcu;
@@ -1295,6 +1296,8 @@ void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp,
 			  struct tcf_proto *tp_head);
 void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc,
 			  struct mini_Qdisc __rcu **p_miniq);
+void mini_qdisc_pair_block_init(struct mini_Qdisc_pair *miniqp,
+				struct tcf_block *block);
 
 static inline int skb_tc_reinsert(struct sk_buff *skb, struct tcf_result *res)
 {

net/core/dev.c

@@ -4848,7 +4848,8 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
 	skb->tc_at_ingress = 1;
 	mini_qdisc_bstats_cpu_update(miniq, skb);
 
-	switch (tcf_classify(skb, miniq->filter_list, &cl_res, false)) {
+	switch (tcf_classify_ingress(skb, miniq->block, miniq->filter_list,
+				     &cl_res, false)) {
 	case TC_ACT_OK:
 	case TC_ACT_RECLASSIFY:
 		skb->tc_index = TC_H_MIN(cl_res.classid);

net/sched/cls_api.c

@@ -22,6 +22,7 @@
 #include <linux/idr.h>
 #include <linux/rhashtable.h>
 #include <linux/jhash.h>
+#include <linux/rculist.h>
 #include <net/net_namespace.h>
 #include <net/sock.h>
 #include <net/netlink.h>
@@ -354,7 +355,7 @@ static struct tcf_chain *tcf_chain_create(struct tcf_block *block,
 	chain = kzalloc(sizeof(*chain), GFP_KERNEL);
 	if (!chain)
 		return NULL;
-	list_add_tail(&chain->list, &block->chain_list);
+	list_add_tail_rcu(&chain->list, &block->chain_list);
 	mutex_init(&chain->filter_chain_lock);
 	chain->block = block;
 	chain->index = chain_index;
@@ -394,7 +395,7 @@ static bool tcf_chain_detach(struct tcf_chain *chain)
 
 	ASSERT_BLOCK_LOCKED(block);
 
-	list_del(&chain->list);
+	list_del_rcu(&chain->list);
 	if (!chain->index)
 		block->chain0.chain = NULL;
 
@@ -453,6 +454,20 @@ static struct tcf_chain *tcf_chain_lookup(struct tcf_block *block,
 	return NULL;
 }
 
+#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
+static struct tcf_chain *tcf_chain_lookup_rcu(const struct tcf_block *block,
+					      u32 chain_index)
+{
+	struct tcf_chain *chain;
+
+	list_for_each_entry_rcu(chain, &block->chain_list, list) {
+		if (chain->index == chain_index)
+			return chain;
+	}
+	return NULL;
+}
+#endif
+
 static int tc_chain_notify(struct tcf_chain *chain, struct sk_buff *oskb,
 			   u32 seq, u16 flags, int event, bool unicast);
 
@@ -1559,12 +1574,15 @@ static int tcf_block_setup(struct tcf_block *block,
  * to this qdisc, (optionally) tests for protocol and asks
  * specific classifiers.
  */
-int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
-		 struct tcf_result *res, bool compat_mode)
+static inline int __tcf_classify(struct sk_buff *skb,
+				 const struct tcf_proto *tp,
+				 const struct tcf_proto *orig_tp,
+				 struct tcf_result *res,
+				 bool compat_mode,
+				 u32 *last_executed_chain)
 {
 #ifdef CONFIG_NET_CLS_ACT
 	const int max_reclassify_loop = 4;
-	const struct tcf_proto *orig_tp = tp;
 	const struct tcf_proto *first_tp;
 	int limit = 0;
 
@@ -1582,21 +1600,11 @@ int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 #ifdef CONFIG_NET_CLS_ACT
 		if (unlikely(err == TC_ACT_RECLASSIFY && !compat_mode)) {
 			first_tp = orig_tp;
+			*last_executed_chain = first_tp->chain->index;
 			goto reset;
 		} else if (unlikely(TC_ACT_EXT_CMP(err, TC_ACT_GOTO_CHAIN))) {
 			first_tp = res->goto_tp;
-
-#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
-			{
-				struct tc_skb_ext *ext;
-
-				ext = skb_ext_add(skb, TC_SKB_EXT);
-				if (WARN_ON_ONCE(!ext))
-					return TC_ACT_SHOT;
-
-				ext->chain = err & TC_ACT_EXT_VAL_MASK;
-			}
-#endif
+			*last_executed_chain = err & TC_ACT_EXT_VAL_MASK;
 			goto reset;
 		}
 #endif
@@ -1619,8 +1627,64 @@ int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
 	goto reclassify;
 #endif
 }
+
+int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
+		 struct tcf_result *res, bool compat_mode)
+{
+	u32 last_executed_chain = 0;
+
+	return __tcf_classify(skb, tp, tp, res, compat_mode,
+			      &last_executed_chain);
+}
 EXPORT_SYMBOL(tcf_classify);
 
+int tcf_classify_ingress(struct sk_buff *skb,
+			 const struct tcf_block *ingress_block,
+			 const struct tcf_proto *tp,
+			 struct tcf_result *res, bool compat_mode)
+{
+#if !IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
+	u32 last_executed_chain = 0;
+
+	return __tcf_classify(skb, tp, tp, res, compat_mode,
+			      &last_executed_chain);
+#else
+	u32 last_executed_chain = tp ? tp->chain->index : 0;
+	const struct tcf_proto *orig_tp = tp;
+	struct tc_skb_ext *ext;
+	int ret;
+
+	ext = skb_ext_find(skb, TC_SKB_EXT);
+
+	if (ext && ext->chain) {
+		struct tcf_chain *fchain;
+
+		fchain = tcf_chain_lookup_rcu(ingress_block, ext->chain);
+		if (!fchain)
+			return TC_ACT_SHOT;
+
+		/* Consume, so cloned/redirect skbs won't inherit ext */
+		skb_ext_del(skb, TC_SKB_EXT);
+
+		tp = rcu_dereference_bh(fchain->filter_chain);
+	}
+
+	ret = __tcf_classify(skb, tp, orig_tp, res, compat_mode,
+			     &last_executed_chain);
+
+	/* If we missed on some chain */
+	if (ret == TC_ACT_UNSPEC && last_executed_chain) {
+		ext = skb_ext_add(skb, TC_SKB_EXT);
+		if (WARN_ON_ONCE(!ext))
+			return TC_ACT_SHOT;
+		ext->chain = last_executed_chain;
+	}
+
+	return ret;
+#endif
+}
+EXPORT_SYMBOL(tcf_classify_ingress);
+
 struct tcf_chain_info {
 	struct tcf_proto __rcu **pprev;
 	struct tcf_proto __rcu *next;

net/sched/sch_generic.c

@@ -1391,6 +1391,14 @@ void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp,
 }
 EXPORT_SYMBOL(mini_qdisc_pair_swap);
 
+void mini_qdisc_pair_block_init(struct mini_Qdisc_pair *miniqp,
+				struct tcf_block *block)
+{
+	miniqp->miniq1.block = block;
+	miniqp->miniq2.block = block;
+}
+EXPORT_SYMBOL(mini_qdisc_pair_block_init);
+
 void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc,
 			  struct mini_Qdisc __rcu **p_miniq)
 {

net/sched/sch_ingress.c

@@ -78,6 +78,7 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt,
 {
 	struct ingress_sched_data *q = qdisc_priv(sch);
 	struct net_device *dev = qdisc_dev(sch);
+	int err;
 
 	net_inc_ingress_queue();
 
@@ -87,7 +88,13 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt,
 	q->block_info.chain_head_change = clsact_chain_head_change;
 	q->block_info.chain_head_change_priv = &q->miniqp;
 
-	return tcf_block_get_ext(&q->block, sch, &q->block_info, extack);
+	err = tcf_block_get_ext(&q->block, sch, &q->block_info, extack);
+	if (err)
+		return err;
+
+	mini_qdisc_pair_block_init(&q->miniqp, q->block);
+
+	return 0;
 }
 
 static void ingress_destroy(struct Qdisc *sch)
@@ -226,6 +233,8 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt,
 	if (err)
 		return err;
 
+	mini_qdisc_pair_block_init(&q->miniqp_ingress, q->ingress_block);
+
 	mini_qdisc_pair_init(&q->miniqp_egress, sch, &dev->miniq_egress);
 
 	q->egress_block_info.binder_type = FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS;