eswitch_offloads.c (revision 1fc5a74b108fc90951890ec513ac81869f5eaff1) - OpenGrok cross reference for /linux/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c

/*
 * Copyright (c) 2016, Mellanox Technologies. All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

#include <linux/etherdevice.h>
#include <linux/idr.h>
#include <linux/mlx5/driver.h>
#include <linux/mlx5/mlx5_ifc.h>
#include <linux/mlx5/vport.h>
#include <linux/mlx5/fs.h>
#include <linux/lockdep.h>
#include "mlx5_core.h"
#include "eswitch.h"
#include "esw/indir_table.h"
#include "esw/acl/ofld.h"
#include "rdma.h"
#include "en.h"
#include "fs_core.h"
#include "lib/mlx5.h"
#include "lib/devcom.h"
#include "lib/sd.h"
#include "lib/eq.h"
#include "lib/fs_chains.h"
#include "en_tc.h"
#include "en/mapping.h"
#include "devlink.h"
#include "lag/lag.h"
#include "en/tc/post_meter.h"
#include "fw_reset.h"

/* There are two match-all miss flows, one for unicast dst mac and
 * one for multicast.
 */
#define MLX5_ESW_MISS_FLOWS (2)
#define UPLINK_REP_INDEX 0

#define MLX5_ESW_VPORT_TBL_SIZE 128
#define MLX5_ESW_VPORT_TBL_NUM_GROUPS  4

#define MLX5_ESW_FT_OFFLOADS_DROP_RULE (1)

#define MLX5_ESW_MAX_CTRL_EQS 4
#define MLX5_ESW_DEFAULT_SF_COMP_EQS 8

static struct esw_vport_tbl_namespace mlx5_esw_vport_tbl_mirror_ns = {
	.max_fte = MLX5_ESW_VPORT_TBL_SIZE,
	.max_num_groups = MLX5_ESW_VPORT_TBL_NUM_GROUPS,
	.flags = 0,
};

static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw,
						     u16 vport_num)
{
	return xa_load(&esw->offloads.vport_reps, vport_num);
}

static void
mlx5_eswitch_set_rule_flow_source(struct mlx5_eswitch *esw,
				  struct mlx5_flow_spec *spec,
				  struct mlx5_esw_flow_attr *attr)
{
	if (!MLX5_CAP_ESW_FLOWTABLE(esw->dev, flow_source) || !attr || !attr->in_rep)
		return;

	if (attr->int_port) {
		spec->flow_context.flow_source = mlx5e_tc_int_port_get_flow_source(attr->int_port);

		return;
	}

	spec->flow_context.flow_source = (attr->in_rep->vport == MLX5_VPORT_UPLINK) ?
					 MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK :
					 MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
}

/* Actually only the upper 16 bits of reg c0 need to be cleared, but the lower 16 bits
 * are not needed as well in the following process. So clear them all for simplicity.
 */
void
mlx5_eswitch_clear_rule_source_port(struct mlx5_eswitch *esw, struct mlx5_flow_spec *spec)
{
	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		void *misc2;

		misc2 = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc2, metadata_reg_c_0, 0);

		misc2 = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc2, metadata_reg_c_0, 0);

		if (!memchr_inv(misc2, 0, MLX5_ST_SZ_BYTES(fte_match_set_misc2)))
			spec->match_criteria_enable &= ~MLX5_MATCH_MISC_PARAMETERS_2;
	}
}

static void
mlx5_eswitch_set_rule_source_port(struct mlx5_eswitch *esw,
				  struct mlx5_flow_spec *spec,
				  struct mlx5_flow_attr *attr,
				  struct mlx5_eswitch *src_esw,
				  u16 vport)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	u32 metadata;
	void *misc2;
	void *misc;

	/* Use metadata matching because vport is not represented by single
	 * VHCA in dual-port RoCE mode, and matching on source vport may fail.
	 */
	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		if (mlx5_esw_indir_table_decap_vport(attr))
			vport = mlx5_esw_indir_table_decap_vport(attr);

		if (!attr->chain && esw_attr && esw_attr->int_port)
			metadata =
				mlx5e_tc_int_port_get_metadata_for_match(esw_attr->int_port);
		else
			metadata =
				mlx5_eswitch_get_vport_metadata_for_match(src_esw, vport);

		misc2 = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc2, metadata_reg_c_0, metadata);

		misc2 = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc2, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_mask());

		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2;
	} else {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
		MLX5_SET(fte_match_set_misc, misc, source_port, vport);

		if (MLX5_CAP_ESW(esw->dev, merged_eswitch))
			MLX5_SET(fte_match_set_misc, misc,
				 source_eswitch_owner_vhca_id,
				 MLX5_CAP_GEN(src_esw->dev, vhca_id));

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
		MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
		if (MLX5_CAP_ESW(esw->dev, merged_eswitch))
			MLX5_SET_TO_ONES(fte_match_set_misc, misc,
					 source_eswitch_owner_vhca_id);

		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS;
	}
}

static int
esw_setup_decap_indir(struct mlx5_eswitch *esw,
		      struct mlx5_flow_attr *attr)
{
	struct mlx5_flow_table *ft;

	if (!(attr->flags & MLX5_ATTR_FLAG_SRC_REWRITE))
		return -EOPNOTSUPP;

	ft = mlx5_esw_indir_table_get(esw, attr,
				      mlx5_esw_indir_table_decap_vport(attr), true);
	return PTR_ERR_OR_ZERO(ft);
}

static void
esw_cleanup_decap_indir(struct mlx5_eswitch *esw,
			struct mlx5_flow_attr *attr)
{
	if (mlx5_esw_indir_table_decap_vport(attr))
		mlx5_esw_indir_table_put(esw,
					 mlx5_esw_indir_table_decap_vport(attr),
					 true);
}

static int
esw_setup_mtu_dest(struct mlx5_flow_destination *dest,
		   struct mlx5e_meter_attr *meter,
		   int i)
{
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_RANGE;
	dest[i].range.field = MLX5_FLOW_DEST_RANGE_FIELD_PKT_LEN;
	dest[i].range.min = 0;
	dest[i].range.max = meter->params.mtu;
	dest[i].range.hit_ft = mlx5e_post_meter_get_mtu_true_ft(meter->post_meter);
	dest[i].range.miss_ft = mlx5e_post_meter_get_mtu_false_ft(meter->post_meter);

	return 0;
}

static int
esw_setup_sampler_dest(struct mlx5_flow_destination *dest,
		       struct mlx5_flow_act *flow_act,
		       u32 sampler_id,
		       int i)
{
	flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_SAMPLER;
	dest[i].sampler_id = sampler_id;

	return 0;
}

static int
esw_setup_ft_dest(struct mlx5_flow_destination *dest,
		  struct mlx5_flow_act *flow_act,
		  struct mlx5_eswitch *esw,
		  struct mlx5_flow_attr *attr,
		  int i)
{
	flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest[i].ft = attr->dest_ft;

	if (mlx5_esw_indir_table_decap_vport(attr))
		return esw_setup_decap_indir(esw, attr);
	return 0;
}

static void
esw_setup_accept_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
		      struct mlx5_fs_chains *chains, int i)
{
	if (mlx5_chains_ignore_flow_level_supported(chains))
		flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest[i].ft = mlx5_chains_get_tc_end_ft(chains);
}

static void
esw_setup_slow_path_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
			 struct mlx5_eswitch *esw, int i)
{
	if (MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level))
		flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest[i].ft = mlx5_eswitch_get_slow_fdb(esw);
}

static int
esw_setup_chain_dest(struct mlx5_flow_destination *dest,
		     struct mlx5_flow_act *flow_act,
		     struct mlx5_fs_chains *chains,
		     u32 chain, u32 prio, u32 level,
		     int i)
{
	struct mlx5_flow_table *ft;

	flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	ft = mlx5_chains_get_table(chains, chain, prio, level);
	if (IS_ERR(ft))
		return PTR_ERR(ft);

	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest[i].ft = ft;
	return  0;
}

static void esw_put_dest_tables_loop(struct mlx5_eswitch *esw, struct mlx5_flow_attr *attr,
				     int from, int to)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);
	int i;

	for (i = from; i < to; i++)
		if (esw_attr->dests[i].flags & MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE)
			mlx5_chains_put_table(chains, 0, 1, 0);
		else if (mlx5_esw_indir_table_needed(esw, attr, esw_attr->dests[i].vport,
						     esw_attr->dests[i].mdev))
			mlx5_esw_indir_table_put(esw, esw_attr->dests[i].vport, false);
}

static bool
esw_is_chain_src_port_rewrite(struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *esw_attr)
{
	int i;

	for (i = esw_attr->split_count; i < esw_attr->out_count; i++)
		if (esw_attr->dests[i].flags & MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE)
			return true;
	return false;
}

static int
esw_setup_chain_src_port_rewrite(struct mlx5_flow_destination *dest,
				 struct mlx5_flow_act *flow_act,
				 struct mlx5_eswitch *esw,
				 struct mlx5_fs_chains *chains,
				 struct mlx5_flow_attr *attr,
				 int *i)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	int err;

	if (!(attr->flags & MLX5_ATTR_FLAG_SRC_REWRITE))
		return -EOPNOTSUPP;

	/* flow steering cannot handle more than one dest with the same ft
	 * in a single flow
	 */
	if (esw_attr->out_count - esw_attr->split_count > 1)
		return -EOPNOTSUPP;

	err = esw_setup_chain_dest(dest, flow_act, chains, attr->dest_chain, 1, 0, *i);
	if (err)
		return err;

	if (esw_attr->dests[esw_attr->split_count].pkt_reformat) {
		flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
		flow_act->pkt_reformat = esw_attr->dests[esw_attr->split_count].pkt_reformat;
	}
	(*i)++;

	return 0;
}

static void esw_cleanup_chain_src_port_rewrite(struct mlx5_eswitch *esw,
					       struct mlx5_flow_attr *attr)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;

	esw_put_dest_tables_loop(esw, attr, esw_attr->split_count, esw_attr->out_count);
}

static bool
esw_is_indir_table(struct mlx5_eswitch *esw, struct mlx5_flow_attr *attr)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	bool result = false;
	int i;

	/* Indirect table is supported only for flows with in_port uplink
	 * and the destination is vport on the same eswitch as the uplink,
	 * return false in case at least one of destinations doesn't meet
	 * this criteria.
	 */
	for (i = esw_attr->split_count; i < esw_attr->out_count; i++) {
		if (esw_attr->dests[i].vport_valid &&
		    mlx5_esw_indir_table_needed(esw, attr, esw_attr->dests[i].vport,
						esw_attr->dests[i].mdev)) {
			result = true;
		} else {
			result = false;
			break;
		}
	}
	return result;
}

static int
esw_setup_indir_table(struct mlx5_flow_destination *dest,
		      struct mlx5_flow_act *flow_act,
		      struct mlx5_eswitch *esw,
		      struct mlx5_flow_attr *attr,
		      int *i)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	int j, err;

	if (!(attr->flags & MLX5_ATTR_FLAG_SRC_REWRITE))
		return -EOPNOTSUPP;

	for (j = esw_attr->split_count; j < esw_attr->out_count; j++, (*i)++) {
		flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
		dest[*i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;

		dest[*i].ft = mlx5_esw_indir_table_get(esw, attr,
						       esw_attr->dests[j].vport, false);
		if (IS_ERR(dest[*i].ft)) {
			err = PTR_ERR(dest[*i].ft);
			goto err_indir_tbl_get;
		}
	}

	if (mlx5_esw_indir_table_decap_vport(attr)) {
		err = esw_setup_decap_indir(esw, attr);
		if (err)
			goto err_indir_tbl_get;
	}

	return 0;

err_indir_tbl_get:
	esw_put_dest_tables_loop(esw, attr, esw_attr->split_count, j);
	return err;
}

static void esw_cleanup_indir_table(struct mlx5_eswitch *esw, struct mlx5_flow_attr *attr)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;

	esw_put_dest_tables_loop(esw, attr, esw_attr->split_count, esw_attr->out_count);
	esw_cleanup_decap_indir(esw, attr);
}

static void
esw_cleanup_chain_dest(struct mlx5_fs_chains *chains, u32 chain, u32 prio, u32 level)
{
	mlx5_chains_put_table(chains, chain, prio, level);
}

static bool esw_same_vhca_id(struct mlx5_core_dev *mdev1, struct mlx5_core_dev *mdev2)
{
	return MLX5_CAP_GEN(mdev1, vhca_id) == MLX5_CAP_GEN(mdev2, vhca_id);
}

static bool esw_setup_uplink_fwd_ipsec_needed(struct mlx5_eswitch *esw,
					      struct mlx5_esw_flow_attr *esw_attr,
					      int attr_idx)
{
	if (esw->offloads.ft_ipsec_tx_pol &&
	    esw_attr->dests[attr_idx].vport_valid &&
	    esw_attr->dests[attr_idx].vport == MLX5_VPORT_UPLINK &&
	    /* To be aligned with software, encryption is needed only for tunnel device */
	    (esw_attr->dests[attr_idx].flags & MLX5_ESW_DEST_ENCAP_VALID) &&
	    esw_attr->dests[attr_idx].vport != esw_attr->in_rep->vport &&
	    esw_same_vhca_id(esw_attr->dests[attr_idx].mdev, esw->dev))
		return true;

	return false;
}

static bool esw_flow_dests_fwd_ipsec_check(struct mlx5_eswitch *esw,
					   struct mlx5_esw_flow_attr *esw_attr)
{
	int i;

	if (!esw->offloads.ft_ipsec_tx_pol)
		return true;

	for (i = 0; i < esw_attr->split_count; i++)
		if (esw_setup_uplink_fwd_ipsec_needed(esw, esw_attr, i))
			return false;

	for (i = esw_attr->split_count; i < esw_attr->out_count; i++)
		if (esw_setup_uplink_fwd_ipsec_needed(esw, esw_attr, i) &&
		    (esw_attr->out_count - esw_attr->split_count > 1))
			return false;

	return true;
}

static void
esw_setup_dest_fwd_vport(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
			 struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *esw_attr,
			 int attr_idx, int dest_idx, bool pkt_reformat)
{
	dest[dest_idx].type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest[dest_idx].vport.num = esw_attr->dests[attr_idx].vport;
	if (MLX5_CAP_ESW(esw->dev, merged_eswitch)) {
		dest[dest_idx].vport.vhca_id =
			MLX5_CAP_GEN(esw_attr->dests[attr_idx].mdev, vhca_id);
		dest[dest_idx].vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID;
		if (dest[dest_idx].vport.num == MLX5_VPORT_UPLINK &&
		    mlx5_lag_is_mpesw(esw->dev))
			dest[dest_idx].type = MLX5_FLOW_DESTINATION_TYPE_UPLINK;
	}
	if (esw_attr->dests[attr_idx].flags & MLX5_ESW_DEST_ENCAP_VALID) {
		if (pkt_reformat) {
			flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
			flow_act->pkt_reformat = esw_attr->dests[attr_idx].pkt_reformat;
		}
		dest[dest_idx].vport.flags |= MLX5_FLOW_DEST_VPORT_REFORMAT_ID;
		dest[dest_idx].vport.pkt_reformat = esw_attr->dests[attr_idx].pkt_reformat;
	}
}

static void
esw_setup_dest_fwd_ipsec(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
			 struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *esw_attr,
			 int attr_idx, int dest_idx, bool pkt_reformat)
{
	dest[dest_idx].ft = esw->offloads.ft_ipsec_tx_pol;
	dest[dest_idx].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	if (pkt_reformat &&
	    esw_attr->dests[attr_idx].flags & MLX5_ESW_DEST_ENCAP_VALID) {
		flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
		flow_act->pkt_reformat = esw_attr->dests[attr_idx].pkt_reformat;
	}
}

static void
esw_setup_vport_dest(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
		     struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *esw_attr,
		     int attr_idx, int dest_idx, bool pkt_reformat)
{
	if (esw_setup_uplink_fwd_ipsec_needed(esw, esw_attr, attr_idx))
		esw_setup_dest_fwd_ipsec(dest, flow_act, esw, esw_attr,
					 attr_idx, dest_idx, pkt_reformat);
	else
		esw_setup_dest_fwd_vport(dest, flow_act, esw, esw_attr,
					 attr_idx, dest_idx, pkt_reformat);
}

static int
esw_setup_vport_dests(struct mlx5_flow_destination *dest, struct mlx5_flow_act *flow_act,
		      struct mlx5_eswitch *esw, struct mlx5_esw_flow_attr *esw_attr,
		      int i)
{
	int j;

	for (j = esw_attr->split_count; j < esw_attr->out_count; j++, i++)
		esw_setup_vport_dest(dest, flow_act, esw, esw_attr, j, i, true);
	return i;
}

static bool
esw_src_port_rewrite_supported(struct mlx5_eswitch *esw)
{
	return MLX5_CAP_GEN(esw->dev, reg_c_preserve) &&
	       mlx5_eswitch_vport_match_metadata_enabled(esw) &&
	       MLX5_CAP_ESW_FLOWTABLE_FDB(esw->dev, ignore_flow_level);
}

static bool
esw_dests_to_int_external(struct mlx5_flow_destination *dests, int max_dest)
{
	bool internal_dest = false, external_dest = false;
	int i;

	for (i = 0; i < max_dest; i++) {
		if (dests[i].type != MLX5_FLOW_DESTINATION_TYPE_VPORT &&
		    dests[i].type != MLX5_FLOW_DESTINATION_TYPE_UPLINK)
			continue;

		/* Uplink dest is external, but considered as internal
		 * if there is reformat because firmware uses LB+hairpin to support it.
		 */
		if (dests[i].vport.num == MLX5_VPORT_UPLINK &&
		    !(dests[i].vport.flags & MLX5_FLOW_DEST_VPORT_REFORMAT_ID))
			external_dest = true;
		else
			internal_dest = true;

		if (internal_dest && external_dest)
			return true;
	}

	return false;
}

static int
esw_setup_dests(struct mlx5_flow_destination *dest,
		struct mlx5_flow_act *flow_act,
		struct mlx5_eswitch *esw,
		struct mlx5_flow_attr *attr,
		struct mlx5_flow_spec *spec,
		int *i)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);
	int err = 0;

	if (!mlx5_eswitch_termtbl_required(esw, attr, flow_act, spec) &&
	    esw_src_port_rewrite_supported(esw))
		attr->flags |= MLX5_ATTR_FLAG_SRC_REWRITE;

	if (attr->flags & MLX5_ATTR_FLAG_SLOW_PATH) {
		esw_setup_slow_path_dest(dest, flow_act, esw, *i);
		(*i)++;
		goto out;
	}

	if (attr->flags & MLX5_ATTR_FLAG_SAMPLE) {
		esw_setup_sampler_dest(dest, flow_act, attr->sample_attr.sampler_id, *i);
		(*i)++;
	} else if (attr->flags & MLX5_ATTR_FLAG_ACCEPT) {
		esw_setup_accept_dest(dest, flow_act, chains, *i);
		(*i)++;
	} else if (attr->flags & MLX5_ATTR_FLAG_MTU) {
		err = esw_setup_mtu_dest(dest, &attr->meter_attr, *i);
		(*i)++;
	} else if (esw_is_indir_table(esw, attr)) {
		err = esw_setup_indir_table(dest, flow_act, esw, attr, i);
	} else if (esw_is_chain_src_port_rewrite(esw, esw_attr)) {
		err = esw_setup_chain_src_port_rewrite(dest, flow_act, esw, chains, attr, i);
	} else {
		*i = esw_setup_vport_dests(dest, flow_act, esw, esw_attr, *i);

		if (attr->dest_ft) {
			err = esw_setup_ft_dest(dest, flow_act, esw, attr, *i);
			(*i)++;
		} else if (attr->dest_chain) {
			err = esw_setup_chain_dest(dest, flow_act, chains, attr->dest_chain,
						   1, 0, *i);
			(*i)++;
		}
	}

	if (attr->extra_split_ft) {
		flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
		dest[*i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
		dest[*i].ft = attr->extra_split_ft;
		(*i)++;
	}

out:
	return err;
}

static void
esw_cleanup_dests(struct mlx5_eswitch *esw,
		  struct mlx5_flow_attr *attr)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);

	if (attr->dest_ft) {
		esw_cleanup_decap_indir(esw, attr);
	} else if (!mlx5e_tc_attr_flags_skip(attr->flags)) {
		if (attr->dest_chain)
			esw_cleanup_chain_dest(chains, attr->dest_chain, 1, 0);
		else if (esw_is_indir_table(esw, attr))
			esw_cleanup_indir_table(esw, attr);
		else if (esw_is_chain_src_port_rewrite(esw, esw_attr))
			esw_cleanup_chain_src_port_rewrite(esw, attr);
	}
}

static void
esw_setup_meter(struct mlx5_flow_attr *attr, struct mlx5_flow_act *flow_act)
{
	struct mlx5e_flow_meter_handle *meter;

	meter = attr->meter_attr.meter;
	flow_act->exe_aso.type = attr->exe_aso_type;
	flow_act->exe_aso.object_id = meter->obj_id;
	flow_act->exe_aso.base_id = mlx5e_flow_meter_get_base_id(meter);
	flow_act->exe_aso.flow_meter.meter_idx = meter->idx;
	flow_act->exe_aso.flow_meter.init_color = MLX5_FLOW_METER_COLOR_GREEN;
	/* use metadata reg 5 for packet color */
	flow_act->exe_aso.return_reg_id = 5;
}

struct mlx5_flow_handle *
mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
				struct mlx5_flow_spec *spec,
				struct mlx5_flow_attr *attr)
{
	struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, };
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);
	bool split = !!(esw_attr->split_count);
	struct mlx5_vport_tbl_attr fwd_attr;
	struct mlx5_flow_destination *dest;
	struct mlx5_flow_handle *rule;
	struct mlx5_flow_table *fdb;
	int i = 0;

	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return ERR_PTR(-EOPNOTSUPP);

	if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
		return ERR_PTR(-EOPNOTSUPP);

	if (!esw_flow_dests_fwd_ipsec_check(esw, esw_attr))
		return ERR_PTR(-EOPNOTSUPP);

	dest = kzalloc_objs(*dest, MLX5_MAX_FLOW_FWD_VPORTS + 1);
	if (!dest)
		return ERR_PTR(-ENOMEM);

	flow_act.action = attr->action;

	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
		flow_act.vlan[0].ethtype = ntohs(esw_attr->vlan_proto[0]);
		flow_act.vlan[0].vid = esw_attr->vlan_vid[0];
		flow_act.vlan[0].prio = esw_attr->vlan_prio[0];
		if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) {
			flow_act.vlan[1].ethtype = ntohs(esw_attr->vlan_proto[1]);
			flow_act.vlan[1].vid = esw_attr->vlan_vid[1];
			flow_act.vlan[1].prio = esw_attr->vlan_prio[1];
		}
	}

	mlx5_eswitch_set_rule_flow_source(esw, spec, esw_attr);

	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
		int err;

		err = esw_setup_dests(dest, &flow_act, esw, attr, spec, &i);
		if (err) {
			rule = ERR_PTR(err);
			goto err_create_goto_table;
		}

		/* Header rewrite with combined wire+loopback in FDB is not allowed */
		if ((flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) &&
		    esw_dests_to_int_external(dest, i)) {
			esw_warn(esw->dev,
				 "FDB: Header rewrite with forwarding to both internal and external dests is not allowed\n");
			rule = ERR_PTR(-EINVAL);
			goto err_esw_get;
		}
	}

	if (esw_attr->decap_pkt_reformat)
		flow_act.pkt_reformat = esw_attr->decap_pkt_reformat;

	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) {
		dest[i].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
		dest[i].counter = attr->counter;
		i++;
	}

	if (attr->outer_match_level != MLX5_MATCH_NONE)
		spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS;
	if (attr->inner_match_level != MLX5_MATCH_NONE)
		spec->match_criteria_enable |= MLX5_MATCH_INNER_HEADERS;

	if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)
		flow_act.modify_hdr = attr->modify_hdr;

	if ((flow_act.action & MLX5_FLOW_CONTEXT_ACTION_EXECUTE_ASO) &&
	    attr->exe_aso_type == MLX5_EXE_ASO_FLOW_METER)
		esw_setup_meter(attr, &flow_act);

	if (split) {
		fwd_attr.chain = attr->chain;
		fwd_attr.prio = attr->prio;
		fwd_attr.vport = esw_attr->in_rep->vport;
		fwd_attr.vport_ns = &mlx5_esw_vport_tbl_mirror_ns;

		fdb = mlx5_esw_vporttbl_get(esw, &fwd_attr);
	} else {
		if (attr->chain || attr->prio)
			fdb = mlx5_chains_get_table(chains, attr->chain,
						    attr->prio, 0);
		else
			fdb = attr->ft;

		if (!(attr->flags & MLX5_ATTR_FLAG_NO_IN_PORT))
			mlx5_eswitch_set_rule_source_port(esw, spec, attr,
							  esw_attr->in_mdev->priv.eswitch,
							  esw_attr->in_rep->vport);
	}
	if (IS_ERR(fdb)) {
		rule = ERR_CAST(fdb);
		goto err_esw_get;
	}

	if (!i) {
		kfree(dest);
		dest = NULL;
	}

	if (mlx5_eswitch_termtbl_required(esw, attr, &flow_act, spec))
		rule = mlx5_eswitch_add_termtbl_rule(esw, fdb, spec, esw_attr,
						     &flow_act, dest, i);
	else
		rule = mlx5_add_flow_rules(fdb, spec, &flow_act, dest, i);
	if (IS_ERR(rule))
		goto err_add_rule;
	else
		atomic64_inc(&esw->offloads.num_flows);

	kfree(dest);
	return rule;

err_add_rule:
	if (split)
		mlx5_esw_vporttbl_put(esw, &fwd_attr);
	else if (attr->chain || attr->prio)
		mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
err_esw_get:
	esw_cleanup_dests(esw, attr);
err_create_goto_table:
	kfree(dest);
	return rule;
}

struct mlx5_flow_handle *
mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
			  struct mlx5_flow_spec *spec,
			  struct mlx5_flow_attr *attr)
{
	struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, };
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);
	struct mlx5_vport_tbl_attr fwd_attr;
	struct mlx5_flow_destination *dest;
	struct mlx5_flow_table *fast_fdb;
	struct mlx5_flow_table *fwd_fdb;
	struct mlx5_flow_handle *rule;
	int i, err = 0;

	dest = kzalloc_objs(*dest, MLX5_MAX_FLOW_FWD_VPORTS + 1);
	if (!dest)
		return ERR_PTR(-ENOMEM);

	fast_fdb = mlx5_chains_get_table(chains, attr->chain, attr->prio, 0);
	if (IS_ERR(fast_fdb)) {
		rule = ERR_CAST(fast_fdb);
		goto err_get_fast;
	}

	fwd_attr.chain = attr->chain;
	fwd_attr.prio = attr->prio;
	fwd_attr.vport = esw_attr->in_rep->vport;
	fwd_attr.vport_ns = &mlx5_esw_vport_tbl_mirror_ns;
	fwd_fdb = mlx5_esw_vporttbl_get(esw, &fwd_attr);
	if (IS_ERR(fwd_fdb)) {
		rule = ERR_CAST(fwd_fdb);
		goto err_get_fwd;
	}

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	for (i = 0; i < esw_attr->split_count; i++) {
		if (esw_attr->dests[i].flags & MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE)
			/* Source port rewrite (forward to ovs internal port or statck device) isn't
			 * supported in the rule of split action.
			 */
			err = -EOPNOTSUPP;
		else
			esw_setup_vport_dest(dest, &flow_act, esw, esw_attr, i, i, false);

		if (err) {
			rule = ERR_PTR(err);
			goto err_chain_src_rewrite;
		}
	}
	dest[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest[i].ft = fwd_fdb;
	i++;

	mlx5_eswitch_set_rule_source_port(esw, spec, attr,
					  esw_attr->in_mdev->priv.eswitch,
					  esw_attr->in_rep->vport);

	if (attr->outer_match_level != MLX5_MATCH_NONE)
		spec->match_criteria_enable |= MLX5_MATCH_OUTER_HEADERS;

	flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL;
	rule = mlx5_add_flow_rules(fast_fdb, spec, &flow_act, dest, i);

	if (IS_ERR(rule)) {
		i = esw_attr->split_count;
		goto err_chain_src_rewrite;
	}

	atomic64_inc(&esw->offloads.num_flows);

	kfree(dest);
	return rule;
err_chain_src_rewrite:
	mlx5_esw_vporttbl_put(esw, &fwd_attr);
err_get_fwd:
	mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
err_get_fast:
	kfree(dest);
	return rule;
}

static void
__mlx5_eswitch_del_rule(struct mlx5_eswitch *esw,
			struct mlx5_flow_handle *rule,
			struct mlx5_flow_attr *attr,
			bool fwd_rule)
{
	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
	struct mlx5_fs_chains *chains = esw_chains(esw);
	bool split = (esw_attr->split_count > 0);
	struct mlx5_vport_tbl_attr fwd_attr;
	int i;

	mlx5_del_flow_rules(rule);

	if (!mlx5e_tc_attr_flags_skip(attr->flags)) {
		/* unref the term table */
		for (i = 0; i < MLX5_MAX_FLOW_FWD_VPORTS; i++) {
			if (esw_attr->dests[i].termtbl)
				mlx5_eswitch_termtbl_put(esw, esw_attr->dests[i].termtbl);
		}
	}

	atomic64_dec(&esw->offloads.num_flows);

	if (fwd_rule || split) {
		fwd_attr.chain = attr->chain;
		fwd_attr.prio = attr->prio;
		fwd_attr.vport = esw_attr->in_rep->vport;
		fwd_attr.vport_ns = &mlx5_esw_vport_tbl_mirror_ns;
	}

	if (fwd_rule)  {
		mlx5_esw_vporttbl_put(esw, &fwd_attr);
		mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
	} else {
		if (split)
			mlx5_esw_vporttbl_put(esw, &fwd_attr);
		else if (attr->chain || attr->prio)
			mlx5_chains_put_table(chains, attr->chain, attr->prio, 0);
		esw_cleanup_dests(esw, attr);
	}
}

void
mlx5_eswitch_del_offloaded_rule(struct mlx5_eswitch *esw,
				struct mlx5_flow_handle *rule,
				struct mlx5_flow_attr *attr)
{
	__mlx5_eswitch_del_rule(esw, rule, attr, false);
}

void
mlx5_eswitch_del_fwd_rule(struct mlx5_eswitch *esw,
			  struct mlx5_flow_handle *rule,
			  struct mlx5_flow_attr *attr)
{
	__mlx5_eswitch_del_rule(esw, rule, attr, true);
}

struct mlx5_flow_handle *
mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *on_esw,
				    struct mlx5_eswitch *from_esw,
				    struct mlx5_eswitch_rep *rep,
				    u32 sqn)
{
	struct mlx5_flow_act flow_act = {0};
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_handle *flow_rule;
	struct mlx5_flow_spec *spec;
	void *misc;
	u16 vport;

	spec = kvzalloc_obj(*spec);
	if (!spec) {
		flow_rule = ERR_PTR(-ENOMEM);
		goto out;
	}

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
	MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn);

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn);

	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;

	/* source vport is the esw manager */
	vport = from_esw->manager_vport;

	if (mlx5_eswitch_vport_match_metadata_enabled(on_esw)) {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_for_match(from_esw, vport));

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_mask());

		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2;
	} else {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
		MLX5_SET(fte_match_set_misc, misc, source_port, vport);

		if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
			MLX5_SET(fte_match_set_misc, misc, source_eswitch_owner_vhca_id,
				 MLX5_CAP_GEN(from_esw->dev, vhca_id));

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
		MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);

		if (MLX5_CAP_ESW(on_esw->dev, merged_eswitch))
			MLX5_SET_TO_ONES(fte_match_set_misc, misc,
					 source_eswitch_owner_vhca_id);

		spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS;
	}

	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest.vport.num = rep->vport;
	dest.vport.vhca_id = MLX5_CAP_GEN(rep->esw->dev, vhca_id);
	dest.vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID;
	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;

	if (rep->vport == MLX5_VPORT_UPLINK &&
	    on_esw == from_esw && on_esw->offloads.ft_ipsec_tx_pol) {
		dest.ft = on_esw->offloads.ft_ipsec_tx_pol;
		flow_act.flags = FLOW_ACT_IGNORE_FLOW_LEVEL;
		dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	} else {
		dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
		dest.vport.num = rep->vport;
		dest.vport.vhca_id = MLX5_CAP_GEN(rep->esw->dev, vhca_id);
		dest.vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID;
	}

	if (MLX5_CAP_ESW_FLOWTABLE(on_esw->dev, flow_source) &&
	    rep->vport == MLX5_VPORT_UPLINK)
		spec->flow_context.flow_source = MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;

	flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(on_esw),
					spec, &flow_act, &dest, 1);
	if (IS_ERR(flow_rule))
		esw_warn(on_esw->dev, "FDB: Failed to add send to vport rule err %pe\n",
			 flow_rule);
out:
	kvfree(spec);
	return flow_rule;
}
EXPORT_SYMBOL(mlx5_eswitch_add_send_to_vport_rule);

void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule)
{
	mlx5_del_flow_rules(rule);
}

void mlx5_eswitch_del_send_to_vport_meta_rule(struct mlx5_flow_handle *rule)
{
	if (rule)
		mlx5_del_flow_rules(rule);
}

struct mlx5_flow_handle *
mlx5_eswitch_add_send_to_vport_meta_rule(struct mlx5_eswitch *esw, u16 vport_num)
{
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_act flow_act = {0};
	struct mlx5_flow_handle *flow_rule;
	struct mlx5_flow_spec *spec;

	spec = kvzalloc_obj(*spec);
	if (!spec)
		return ERR_PTR(-ENOMEM);

	MLX5_SET(fte_match_param, spec->match_criteria,
		 misc_parameters_2.metadata_reg_c_0, mlx5_eswitch_get_vport_metadata_mask());
	MLX5_SET(fte_match_param, spec->match_criteria,
		 misc_parameters_2.metadata_reg_c_1, ESW_TUN_MASK);
	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_c_1,
		 ESW_TUN_SLOW_TABLE_GOTO_VPORT_MARK);

	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;

	MLX5_SET(fte_match_param, spec->match_value, misc_parameters_2.metadata_reg_c_0,
		 mlx5_eswitch_get_vport_metadata_for_match(esw, vport_num));
	dest.vport.num = vport_num;

	flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					spec, &flow_act, &dest, 1);
	if (IS_ERR(flow_rule))
		esw_warn(esw->dev, "FDB: Failed to add send to vport meta rule vport %d, err %pe\n",
			 vport_num, flow_rule);

	kvfree(spec);
	return flow_rule;
}

static bool mlx5_eswitch_reg_c1_loopback_supported(struct mlx5_eswitch *esw)
{
	return MLX5_CAP_ESW_FLOWTABLE(esw->dev, fdb_to_vport_reg_c_id) &
	       MLX5_FDB_TO_VPORT_REG_C_1;
}

static int esw_set_passing_vport_metadata(struct mlx5_eswitch *esw, bool enable)
{
	u32 out[MLX5_ST_SZ_DW(query_esw_vport_context_out)] = {};
	u32 min[MLX5_ST_SZ_DW(modify_esw_vport_context_in)] = {};
	u32 in[MLX5_ST_SZ_DW(query_esw_vport_context_in)] = {};
	u8 curr, wanted;
	int err;

	if (!mlx5_eswitch_reg_c1_loopback_supported(esw) &&
	    !mlx5_eswitch_vport_match_metadata_enabled(esw))
		return 0;

	MLX5_SET(query_esw_vport_context_in, in, opcode,
		 MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT);
	err = mlx5_cmd_exec_inout(esw->dev, query_esw_vport_context, in, out);
	if (err)
		return err;

	curr = MLX5_GET(query_esw_vport_context_out, out,
			esw_vport_context.fdb_to_vport_reg_c_id);
	wanted = MLX5_FDB_TO_VPORT_REG_C_0;
	if (mlx5_eswitch_reg_c1_loopback_supported(esw))
		wanted |= MLX5_FDB_TO_VPORT_REG_C_1;

	if (enable)
		curr |= wanted;
	else
		curr &= ~wanted;

	MLX5_SET(modify_esw_vport_context_in, min,
		 esw_vport_context.fdb_to_vport_reg_c_id, curr);
	MLX5_SET(modify_esw_vport_context_in, min,
		 field_select.fdb_to_vport_reg_c_id, 1);

	err = mlx5_eswitch_modify_esw_vport_context(esw->dev, 0, false, min);
	if (!err) {
		if (enable && (curr & MLX5_FDB_TO_VPORT_REG_C_1))
			esw->flags |= MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED;
		else
			esw->flags &= ~MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED;
	}

	return err;
}

static void peer_miss_rules_setup(struct mlx5_eswitch *esw,
				  struct mlx5_core_dev *peer_dev,
				  struct mlx5_flow_spec *spec,
				  struct mlx5_flow_destination *dest)
{
	void *misc;

	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
				    misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_mask());

		spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
	} else {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
				    misc_parameters);

		MLX5_SET(fte_match_set_misc, misc, source_eswitch_owner_vhca_id,
			 MLX5_CAP_GEN(peer_dev, vhca_id));

		spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
				    misc_parameters);
		MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
		MLX5_SET_TO_ONES(fte_match_set_misc, misc,
				 source_eswitch_owner_vhca_id);
	}

	dest->type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest->vport.num = peer_dev->priv.eswitch->manager_vport;
	dest->vport.vhca_id = MLX5_CAP_GEN(peer_dev, vhca_id);
	dest->vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID;
}

static void esw_set_peer_miss_rule_source_port(struct mlx5_eswitch *esw,
					       struct mlx5_eswitch *peer_esw,
					       struct mlx5_flow_spec *spec,
					       u16 vport)
{
	void *misc;

	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
				    misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_for_match(peer_esw,
								   vport));
	} else {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
				    misc_parameters);
		MLX5_SET(fte_match_set_misc, misc, source_port, vport);
	}
}

static int esw_add_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
				       struct mlx5_core_dev *peer_dev)
{
	struct mlx5_eswitch *peer_esw = peer_dev->priv.eswitch;
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_act flow_act = {0};
	struct mlx5_flow_handle **flows;
	struct mlx5_flow_handle *flow;
	struct mlx5_vport *peer_vport;
	struct mlx5_flow_spec *spec;
	int err;
	unsigned long i;
	void *misc;

	if (!MLX5_VPORT_MANAGER(peer_dev) &&
	    !mlx5_core_is_ecpf_esw_manager(peer_dev))
		return 0;

	spec = kvzalloc_obj(*spec);
	if (!spec)
		return -ENOMEM;

	peer_miss_rules_setup(esw, peer_dev, spec, &dest);

	flows = kvzalloc_objs(*flows, peer_esw->total_vports);
	if (!flows) {
		err = -ENOMEM;
		goto alloc_flows_err;
	}

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
			    misc_parameters);

	if (mlx5_core_is_ecpf_esw_manager(peer_dev) &&
	    mlx5_esw_host_functions_enabled(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw,
						    MLX5_VPORT_HOST_PF);
		esw_set_peer_miss_rule_source_port(esw, peer_esw, spec,
						   MLX5_VPORT_HOST_PF);

		flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					   spec, &flow_act, &dest, 1);
		if (IS_ERR(flow)) {
			err = PTR_ERR(flow);
			goto add_pf_flow_err;
		}
		flows[peer_vport->index] = flow;
	}

	mlx5_esw_for_each_spf_vport(peer_esw, i, peer_vport) {
		esw_set_peer_miss_rule_source_port(esw, peer_esw, spec,
						   peer_vport->vport);

		flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					   spec, &flow_act, &dest, 1);
		if (IS_ERR(flow)) {
			err = PTR_ERR(flow);
			goto add_ecpf_flow_err;
		}
		flows[peer_vport->index] = flow;
	}

	if (mlx5_ecpf_vport_exists(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw, MLX5_VPORT_ECPF);
		MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_ECPF);
		flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					   spec, &flow_act, &dest, 1);
		if (IS_ERR(flow)) {
			err = PTR_ERR(flow);
			goto add_ecpf_flow_err;
		}
		flows[peer_vport->index] = flow;
	}

	mlx5_esw_for_each_vf_vport(peer_esw, i, peer_vport,
				   mlx5_core_max_vfs(peer_dev)) {
		esw_set_peer_miss_rule_source_port(esw, peer_esw, spec,
						   peer_vport->vport);
		flow = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					   spec, &flow_act, &dest, 1);
		if (IS_ERR(flow)) {
			err = PTR_ERR(flow);
			goto add_vf_flow_err;
		}
		flows[peer_vport->index] = flow;
	}

	if (mlx5_core_ec_sriov_enabled(peer_dev)) {
		mlx5_esw_for_each_ec_vf_vport(peer_esw, i, peer_vport,
					      mlx5_core_max_ec_vfs(peer_dev)) {
			esw_set_peer_miss_rule_source_port(esw, peer_esw,
							   spec,
							   peer_vport->vport);
			flow = mlx5_add_flow_rules(esw->fdb_table.offloads.slow_fdb,
						   spec, &flow_act, &dest, 1);
			if (IS_ERR(flow)) {
				err = PTR_ERR(flow);
				goto add_ec_vf_flow_err;
			}
			flows[peer_vport->index] = flow;
		}
	}

	err = xa_insert(&esw->fdb_table.offloads.peer_miss_rules,
			MLX5_CAP_GEN(peer_dev, vhca_id), flows, GFP_KERNEL);
	if (err)
		goto add_ec_vf_flow_err;

	kvfree(spec);
	return 0;

add_ec_vf_flow_err:
	mlx5_esw_for_each_ec_vf_vport(peer_esw, i, peer_vport,
				      mlx5_core_max_ec_vfs(peer_dev)) {
		if (!flows[peer_vport->index])
			continue;
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}
add_vf_flow_err:
	mlx5_esw_for_each_vf_vport(peer_esw, i, peer_vport,
				   mlx5_core_max_vfs(peer_dev)) {
		if (!flows[peer_vport->index])
			continue;
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}
	if (mlx5_ecpf_vport_exists(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw, MLX5_VPORT_ECPF);
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}
add_ecpf_flow_err:
	mlx5_esw_for_each_spf_vport(peer_esw, i, peer_vport) {
		if (!flows[peer_vport->index])
			continue;
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}
	if (mlx5_core_is_ecpf_esw_manager(peer_dev) &&
	    mlx5_esw_host_functions_enabled(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw,
						    MLX5_VPORT_HOST_PF);
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}
add_pf_flow_err:
	esw_warn(esw->dev, "FDB: Failed to add peer miss flow rule err %d\n", err);
	kvfree(flows);
alloc_flows_err:
	kvfree(spec);
	return err;
}

static void esw_del_fdb_peer_miss_rules(struct mlx5_eswitch *esw,
					struct mlx5_core_dev *peer_dev)
{
	struct mlx5_eswitch *peer_esw = peer_dev->priv.eswitch;
	u16 peer_vhca_id = MLX5_CAP_GEN(peer_dev, vhca_id);
	struct mlx5_flow_handle **flows;
	struct mlx5_vport *peer_vport;
	unsigned long i;

	flows = xa_erase(&esw->fdb_table.offloads.peer_miss_rules,
			 peer_vhca_id);
	if (!flows)
		return;

	if (mlx5_core_ec_sriov_enabled(peer_dev)) {
		mlx5_esw_for_each_ec_vf_vport(peer_esw, i, peer_vport,
					      mlx5_core_max_ec_vfs(peer_dev))
			mlx5_del_flow_rules(flows[peer_vport->index]);
	}

	mlx5_esw_for_each_vf_vport(peer_esw, i, peer_vport,
				   mlx5_core_max_vfs(peer_dev))
		mlx5_del_flow_rules(flows[peer_vport->index]);

	if (mlx5_ecpf_vport_exists(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw, MLX5_VPORT_ECPF);
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}

	mlx5_esw_for_each_spf_vport(peer_esw, i, peer_vport)
		mlx5_del_flow_rules(flows[peer_vport->index]);

	if (mlx5_core_is_ecpf_esw_manager(peer_dev) &&
	    mlx5_esw_host_functions_enabled(peer_dev)) {
		peer_vport = mlx5_eswitch_get_vport(peer_esw,
						    MLX5_VPORT_HOST_PF);
		mlx5_del_flow_rules(flows[peer_vport->index]);
	}

	kvfree(flows);
}

static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw)
{
	struct mlx5_flow_act flow_act = {0};
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_handle *flow_rule = NULL;
	struct mlx5_flow_spec *spec;
	void *headers_c;
	void *headers_v;
	int err = 0;
	u8 *dmac_c;
	u8 *dmac_v;

	spec = kvzalloc_obj(*spec);
	if (!spec) {
		err = -ENOMEM;
		goto out;
	}

	spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
	headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
				 outer_headers);
	dmac_c = MLX5_ADDR_OF(fte_match_param, headers_c,
			      outer_headers.dmac_47_16);
	dmac_c[0] = 0x01;

	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest.vport.num = esw->manager_vport;
	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;

	flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					spec, &flow_act, &dest, 1);
	if (IS_ERR(flow_rule)) {
		err = PTR_ERR(flow_rule);
		esw_warn(esw->dev,  "FDB: Failed to add unicast miss flow rule err %d\n", err);
		goto out;
	}

	esw->fdb_table.offloads.miss_rule_uni = flow_rule;

	headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
				 outer_headers);
	dmac_v = MLX5_ADDR_OF(fte_match_param, headers_v,
			      outer_headers.dmac_47_16);
	dmac_v[0] = 0x01;
	flow_rule = mlx5_add_flow_rules(mlx5_eswitch_get_slow_fdb(esw),
					spec, &flow_act, &dest, 1);
	if (IS_ERR(flow_rule)) {
		err = PTR_ERR(flow_rule);
		esw_warn(esw->dev, "FDB: Failed to add multicast miss flow rule err %d\n", err);
		mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_uni);
		goto out;
	}

	esw->fdb_table.offloads.miss_rule_multi = flow_rule;

out:
	kvfree(spec);
	return err;
}

struct mlx5_flow_handle *
esw_add_restore_rule(struct mlx5_eswitch *esw, u32 tag)
{
	struct mlx5_flow_act flow_act = { .flags = FLOW_ACT_NO_APPEND, };
	struct mlx5_flow_table *ft = esw->offloads.ft_offloads_restore;
	struct mlx5_flow_context *flow_context;
	struct mlx5_flow_handle *flow_rule;
	struct mlx5_flow_destination dest;
	struct mlx5_flow_spec *spec;
	void *misc;

	if (!mlx5_eswitch_reg_c1_loopback_supported(esw))
		return ERR_PTR(-EOPNOTSUPP);

	spec = kvzalloc_obj(*spec);
	if (!spec)
		return ERR_PTR(-ENOMEM);

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
			    misc_parameters_2);
	MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
		 ESW_REG_C0_USER_DATA_METADATA_MASK);
	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
			    misc_parameters_2);
	MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0, tag);
	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST |
			  MLX5_FLOW_CONTEXT_ACTION_MOD_HDR;
	flow_act.modify_hdr = esw->offloads.restore_copy_hdr_id;

	flow_context = &spec->flow_context;
	flow_context->flags |= FLOW_CONTEXT_HAS_TAG;
	flow_context->flow_tag = tag;
	dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
	dest.ft = esw->offloads.ft_offloads;

	flow_rule = mlx5_add_flow_rules(ft, spec, &flow_act, &dest, 1);
	kvfree(spec);

	if (IS_ERR(flow_rule))
		esw_warn(esw->dev,
			 "Failed to create restore rule for tag: %d, err(%d)\n",
			 tag, (int)PTR_ERR(flow_rule));

	return flow_rule;
}

struct mlx5_flow_group *
mlx5_esw_lag_demux_fg_create(struct mlx5_eswitch *esw,
			     struct mlx5_flow_table *ft)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *fg;
	void *match_criteria;
	void *flow_group_in;

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw))
		return ERR_PTR(-EOPNOTSUPP);

	if (IS_ERR(ft))
		return ERR_CAST(ft);

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return ERR_PTR(-ENOMEM);

	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in,
				      match_criteria);
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_MISC_PARAMETERS_2);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
		 ft->max_fte - 1);

	MLX5_SET(fte_match_param, match_criteria,
		 misc_parameters_2.metadata_reg_c_0,
		 mlx5_eswitch_get_vport_metadata_mask());

	fg = mlx5_create_flow_group(ft, flow_group_in);
	kvfree(flow_group_in);
	if (IS_ERR(fg))
		esw_warn(esw->dev, "Can't create LAG demux flow group\n");

	return fg;
}

struct mlx5_flow_handle *
mlx5_esw_lag_demux_rule_create(struct mlx5_eswitch *esw, u16 vport_num,
			       struct mlx5_flow_table *lag_ft)
{
	struct mlx5_flow_spec *spec = kvzalloc_obj(*spec);
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_act flow_act = {};
	struct mlx5_flow_handle *ret;
	void *misc;

	if (!spec)
		return ERR_PTR(-ENOMEM);

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		kvfree(spec);
		return ERR_PTR(-EOPNOTSUPP);
	}

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
			    misc_parameters_2);
	MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
		 mlx5_eswitch_get_vport_metadata_mask());
	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
			    misc_parameters_2);
	MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
		 mlx5_eswitch_get_vport_metadata_for_match(esw, vport_num));

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	dest.type = MLX5_FLOW_DESTINATION_TYPE_VHCA_RX;
	dest.vhca.id = MLX5_CAP_GEN(esw->dev, vhca_id);

	ret = mlx5_add_flow_rules(lag_ft, spec, &flow_act, &dest, 1);
	kvfree(spec);
	return ret;
}

#define MAX_PF_SQ 256
#define MAX_SQ_NVPORTS 32

void
mlx5_esw_set_flow_group_source_port(struct mlx5_eswitch *esw,
				    u32 *flow_group_in,
				    int match_params)
{
	void *match_criteria = MLX5_ADDR_OF(create_flow_group_in,
					    flow_group_in,
					    match_criteria);

	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		MLX5_SET(create_flow_group_in, flow_group_in,
			 match_criteria_enable,
			 MLX5_MATCH_MISC_PARAMETERS_2 | match_params);

		MLX5_SET(fte_match_param, match_criteria,
			 misc_parameters_2.metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_mask());
	} else {
		MLX5_SET(create_flow_group_in, flow_group_in,
			 match_criteria_enable,
			 MLX5_MATCH_MISC_PARAMETERS | match_params);

		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
				 misc_parameters.source_port);
	}
}

#if IS_ENABLED(CONFIG_MLX5_CLS_ACT)
static void esw_vport_tbl_put(struct mlx5_eswitch *esw)
{
	struct mlx5_vport_tbl_attr attr;
	struct mlx5_vport *vport;
	unsigned long i;

	attr.chain = 0;
	attr.prio = 1;
	mlx5_esw_for_each_vport(esw, i, vport) {
		attr.vport = vport->vport;
		attr.vport_ns = &mlx5_esw_vport_tbl_mirror_ns;
		mlx5_esw_vporttbl_put(esw, &attr);
	}
}

static int esw_vport_tbl_get(struct mlx5_eswitch *esw)
{
	struct mlx5_vport_tbl_attr attr;
	struct mlx5_flow_table *fdb;
	struct mlx5_vport *vport;
	unsigned long i;

	attr.chain = 0;
	attr.prio = 1;
	mlx5_esw_for_each_vport(esw, i, vport) {
		attr.vport = vport->vport;
		attr.vport_ns = &mlx5_esw_vport_tbl_mirror_ns;
		fdb = mlx5_esw_vporttbl_get(esw, &attr);
		if (IS_ERR(fdb))
			goto out;
	}
	return 0;

out:
	esw_vport_tbl_put(esw);
	return PTR_ERR(fdb);
}

#define fdb_modify_header_fwd_to_table_supported(esw) \
	(MLX5_CAP_ESW_FLOWTABLE((esw)->dev, fdb_modify_header_fwd_to_table))
static void esw_init_chains_offload_flags(struct mlx5_eswitch *esw, u32 *flags)
{
	struct mlx5_core_dev *dev = esw->dev;

	if (MLX5_CAP_ESW_FLOWTABLE_FDB(dev, ignore_flow_level))
		*flags |= MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED;

	if (!MLX5_CAP_ESW_FLOWTABLE(dev, multi_fdb_encap) &&
	    esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE) {
		*flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED;
		esw_warn(dev, "Tc chains and priorities offload aren't supported, update firmware if needed\n");
	} else if (!mlx5_eswitch_reg_c1_loopback_enabled(esw)) {
		*flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED;
		esw_warn(dev, "Tc chains and priorities offload aren't supported\n");
	} else if (!fdb_modify_header_fwd_to_table_supported(esw)) {
		/* Disabled when ttl workaround is needed, e.g
		 * when ESWITCH_IPV4_TTL_MODIFY_ENABLE = true in mlxconfig
		 */
		esw_warn(dev,
			 "Tc chains and priorities offload aren't supported, check firmware version, or mlxconfig settings\n");
		*flags &= ~MLX5_CHAINS_AND_PRIOS_SUPPORTED;
	} else {
		*flags |= MLX5_CHAINS_AND_PRIOS_SUPPORTED;
		esw_info(dev, "Supported tc chains and prios offload\n");
	}

	if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE)
		*flags |= MLX5_CHAINS_FT_TUNNEL_SUPPORTED;
}

static int
esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb)
{
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_table *nf_ft, *ft;
	struct mlx5_chains_attr attr = {};
	struct mlx5_fs_chains *chains;
	int err;

	esw_init_chains_offload_flags(esw, &attr.flags);
	attr.ns = MLX5_FLOW_NAMESPACE_FDB;
	attr.max_grp_num = esw->params.large_group_num;
	attr.default_ft = miss_fdb;
	attr.mapping = esw->offloads.reg_c0_obj_pool;
	attr.fs_base_prio = FDB_BYPASS_PATH;

	chains = mlx5_chains_create(dev, &attr);
	if (IS_ERR(chains)) {
		err = PTR_ERR(chains);
		esw_warn(dev, "Failed to create fdb chains err(%d)\n", err);
		return err;
	}
	mlx5_chains_print_info(chains);

	esw->fdb_table.offloads.esw_chains_priv = chains;

	/* Create tc_end_ft which is the always created ft chain */
	nf_ft = mlx5_chains_get_table(chains, mlx5_chains_get_nf_ft_chain(chains),
				      1, 0);
	if (IS_ERR(nf_ft)) {
		err = PTR_ERR(nf_ft);
		goto nf_ft_err;
	}

	/* Always open the root for fast path */
	ft = mlx5_chains_get_table(chains, 0, 1, 0);
	if (IS_ERR(ft)) {
		err = PTR_ERR(ft);
		goto level_0_err;
	}

	/* Open level 1 for split fdb rules now if prios isn't supported  */
	if (!mlx5_chains_prios_supported(chains)) {
		err = esw_vport_tbl_get(esw);
		if (err)
			goto level_1_err;
	}

	mlx5_chains_set_end_ft(chains, nf_ft);

	return 0;

level_1_err:
	mlx5_chains_put_table(chains, 0, 1, 0);
level_0_err:
	mlx5_chains_put_table(chains, mlx5_chains_get_nf_ft_chain(chains), 1, 0);
nf_ft_err:
	mlx5_chains_destroy(chains);
	esw->fdb_table.offloads.esw_chains_priv = NULL;

	return err;
}

static void
esw_chains_destroy(struct mlx5_eswitch *esw, struct mlx5_fs_chains *chains)
{
	if (!mlx5_chains_prios_supported(chains))
		esw_vport_tbl_put(esw);
	mlx5_chains_put_table(chains, 0, 1, 0);
	mlx5_chains_put_table(chains, mlx5_chains_get_nf_ft_chain(chains), 1, 0);
	mlx5_chains_destroy(chains);
}

#else /* CONFIG_MLX5_CLS_ACT */

static int
esw_chains_create(struct mlx5_eswitch *esw, struct mlx5_flow_table *miss_fdb)
{ return 0; }

static void
esw_chains_destroy(struct mlx5_eswitch *esw, struct mlx5_fs_chains *chains)
{}

#endif

static int
esw_create_send_to_vport_group(struct mlx5_eswitch *esw,
			       struct mlx5_flow_table *fdb,
			       u32 *flow_group_in,
			       int *ix)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	void *match_criteria;
	int count, err = 0;

	memset(flow_group_in, 0, inlen);

	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, MLX5_MATCH_MISC_PARAMETERS);

	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);
	MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_sqn);

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw) &&
	    MLX5_CAP_ESW(esw->dev, merged_eswitch)) {
		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
				 misc_parameters.source_eswitch_owner_vhca_id);
		MLX5_SET(create_flow_group_in, flow_group_in,
			 source_eswitch_owner_vhca_id_valid, 1);
	}

	/* See comment at table_size calculation */
	count = MLX5_MAX_PORTS * (esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, *ix + count - 1);
	*ix += count;

	g = mlx5_create_flow_group(fdb, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev, "Failed to create send-to-vport flow group err(%d)\n", err);
		goto out;
	}
	esw->fdb_table.offloads.send_to_vport_grp = g;

out:
	return err;
}

static int
esw_create_meta_send_to_vport_group(struct mlx5_eswitch *esw,
				    struct mlx5_flow_table *fdb,
				    u32 *flow_group_in,
				    int *ix)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	void *match_criteria;
	int err = 0;

	if (!esw_src_port_rewrite_supported(esw))
		return 0;

	memset(flow_group_in, 0, inlen);

	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_MISC_PARAMETERS_2);

	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria);

	MLX5_SET(fte_match_param, match_criteria,
		 misc_parameters_2.metadata_reg_c_0,
		 mlx5_eswitch_get_vport_metadata_mask());
	MLX5_SET(fte_match_param, match_criteria,
		 misc_parameters_2.metadata_reg_c_1, ESW_TUN_MASK);

	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, *ix);
	MLX5_SET(create_flow_group_in, flow_group_in,
		 end_flow_index, *ix + esw->total_vports - 1);
	*ix += esw->total_vports;

	g = mlx5_create_flow_group(fdb, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev,
			 "Failed to create send-to-vport meta flow group err(%d)\n", err);
		goto send_vport_meta_err;
	}
	esw->fdb_table.offloads.send_to_vport_meta_grp = g;

	return 0;

send_vport_meta_err:
	return err;
}

static int
esw_create_peer_esw_miss_group(struct mlx5_eswitch *esw,
			       struct mlx5_flow_table *fdb,
			       u32 *flow_group_in,
			       int *ix)
{
	int max_peer_ports = (esw->total_vports - 1) * (MLX5_MAX_PORTS - 1);
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	void *match_criteria;
	int err = 0;

	if (!MLX5_CAP_ESW(esw->dev, merged_eswitch))
		return 0;

	memset(flow_group_in, 0, inlen);

	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, 0);

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		match_criteria = MLX5_ADDR_OF(create_flow_group_in,
					      flow_group_in,
					      match_criteria);

		MLX5_SET_TO_ONES(fte_match_param, match_criteria,
				 misc_parameters.source_eswitch_owner_vhca_id);

		MLX5_SET(create_flow_group_in, flow_group_in,
			 source_eswitch_owner_vhca_id_valid, 1);
	}

	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, *ix);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
		 *ix + max_peer_ports);
	*ix += max_peer_ports + 1;

	g = mlx5_create_flow_group(fdb, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev, "Failed to create peer miss flow group err(%d)\n", err);
		goto out;
	}
	esw->fdb_table.offloads.peer_miss_grp = g;

out:
	return err;
}

static int
esw_create_miss_group(struct mlx5_eswitch *esw,
		      struct mlx5_flow_table *fdb,
		      u32 *flow_group_in,
		      int *ix)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	void *match_criteria;
	int err = 0;
	u8 *dmac;

	memset(flow_group_in, 0, inlen);

	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_OUTER_HEADERS);
	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in,
				      match_criteria);
	dmac = MLX5_ADDR_OF(fte_match_param, match_criteria,
			    outer_headers.dmac_47_16);
	dmac[0] = 0x01;

	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, *ix);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
		 *ix + MLX5_ESW_MISS_FLOWS);

	g = mlx5_create_flow_group(fdb, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev, "Failed to create miss flow group err(%d)\n", err);
		goto miss_err;
	}
	esw->fdb_table.offloads.miss_grp = g;

	err = esw_add_fdb_miss_rule(esw);
	if (err)
		goto miss_rule_err;

	return 0;

miss_rule_err:
	mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp);
miss_err:
	return err;
}

static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_table_attr ft_attr = {};
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_namespace *root_ns;
	struct mlx5_flow_table *fdb = NULL;
	int table_size, ix = 0, err = 0;
	u32 flags = 0, *flow_group_in;

	esw_debug(esw->dev, "Create offloads FDB Tables\n");

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return -ENOMEM;

	root_ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB);
	if (!root_ns) {
		esw_warn(dev, "Failed to get FDB flow namespace\n");
		err = -EOPNOTSUPP;
		goto ns_err;
	}
	esw->fdb_table.offloads.ns = root_ns;
	err = mlx5_flow_namespace_set_mode(root_ns,
					   esw->dev->priv.steering->mode);
	if (err) {
		esw_warn(dev, "Failed to set FDB namespace steering mode\n");
		goto ns_err;
	}

	/* To be strictly correct:
	 *	MLX5_MAX_PORTS * (esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ)
	 * should be:
	 *	esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ +
	 *	peer_esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ
	 * but as the peer device might not be in switchdev mode it's not
	 * possible. We use the fact that by default FW sets max vfs and max sfs
	 * to the same value on both devices. If it needs to be changed in the future note
	 * the peer miss group should also be created based on the number of
	 * total vports of the peer (currently is also uses esw->total_vports).
	 */
	table_size = MLX5_MAX_PORTS * (esw->total_vports * MAX_SQ_NVPORTS + MAX_PF_SQ) +
		     esw->total_vports * MLX5_MAX_PORTS + MLX5_ESW_MISS_FLOWS;

	/* create the slow path fdb with encap set, so further table instances
	 * can be created at run time while VFs are probed if the FW allows that.
	 */
	if (esw->offloads.encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE)
		flags |= (MLX5_FLOW_TABLE_TUNNEL_EN_REFORMAT |
			  MLX5_FLOW_TABLE_TUNNEL_EN_DECAP);

	ft_attr.flags = flags;
	ft_attr.max_fte = table_size;
	ft_attr.prio = FDB_SLOW_PATH;

	fdb = mlx5_create_flow_table(root_ns, &ft_attr);
	if (IS_ERR(fdb)) {
		err = PTR_ERR(fdb);
		esw_warn(dev, "Failed to create slow path FDB Table err %d\n", err);
		goto slow_fdb_err;
	}
	esw->fdb_table.offloads.slow_fdb = fdb;

	/* Create empty TC-miss managed table. This allows plugging in following
	 * priorities without directly exposing their level 0 table to
	 * eswitch_offloads and passing it as miss_fdb to following call to
	 * esw_chains_create().
	 */
	memset(&ft_attr, 0, sizeof(ft_attr));
	ft_attr.prio = FDB_TC_MISS;
	esw->fdb_table.offloads.tc_miss_table = mlx5_create_flow_table(root_ns, &ft_attr);
	if (IS_ERR(esw->fdb_table.offloads.tc_miss_table)) {
		err = PTR_ERR(esw->fdb_table.offloads.tc_miss_table);
		esw_warn(dev, "Failed to create TC miss FDB Table err %d\n", err);
		goto tc_miss_table_err;
	}

	err = esw_chains_create(esw, esw->fdb_table.offloads.tc_miss_table);
	if (err) {
		esw_warn(dev, "Failed to open fdb chains err(%d)\n", err);
		goto fdb_chains_err;
	}

	err = esw_create_send_to_vport_group(esw, fdb, flow_group_in, &ix);
	if (err)
		goto send_vport_err;

	err = esw_create_meta_send_to_vport_group(esw, fdb, flow_group_in, &ix);
	if (err)
		goto send_vport_meta_err;

	err = esw_create_peer_esw_miss_group(esw, fdb, flow_group_in, &ix);
	if (err)
		goto peer_miss_err;

	err = esw_create_miss_group(esw, fdb, flow_group_in, &ix);
	if (err)
		goto miss_err;

	kvfree(flow_group_in);
	return 0;

miss_err:
	if (MLX5_CAP_ESW(esw->dev, merged_eswitch))
		mlx5_destroy_flow_group(esw->fdb_table.offloads.peer_miss_grp);
peer_miss_err:
	if (esw->fdb_table.offloads.send_to_vport_meta_grp)
		mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_meta_grp);
send_vport_meta_err:
	mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp);
send_vport_err:
	esw_chains_destroy(esw, esw_chains(esw));
fdb_chains_err:
	mlx5_destroy_flow_table(esw->fdb_table.offloads.tc_miss_table);
tc_miss_table_err:
	mlx5_destroy_flow_table(mlx5_eswitch_get_slow_fdb(esw));
slow_fdb_err:
	/* Holds true only as long as DMFS is the default */
	mlx5_flow_namespace_set_mode(root_ns, MLX5_FLOW_STEERING_MODE_DMFS);
ns_err:
	kvfree(flow_group_in);
	return err;
}

static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw)
{
	if (!mlx5_eswitch_get_slow_fdb(esw))
		return;

	esw_debug(esw->dev, "Destroy offloads FDB Tables\n");
	mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_multi);
	mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_uni);
	mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp);
	if (esw->fdb_table.offloads.send_to_vport_meta_grp)
		mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_meta_grp);
	if (MLX5_CAP_ESW(esw->dev, merged_eswitch))
		mlx5_destroy_flow_group(esw->fdb_table.offloads.peer_miss_grp);
	mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp);

	esw_chains_destroy(esw, esw_chains(esw));

	mlx5_destroy_flow_table(esw->fdb_table.offloads.tc_miss_table);
	mlx5_destroy_flow_table(mlx5_eswitch_get_slow_fdb(esw));
	/* Holds true only as long as DMFS is the default */
	mlx5_flow_namespace_set_mode(esw->fdb_table.offloads.ns,
				     MLX5_FLOW_STEERING_MODE_DMFS);
}

static int esw_get_nr_ft_offloads_steering_src_ports(struct mlx5_eswitch *esw)
{
	int nvports;

	nvports = esw->total_vports + MLX5_ESW_MISS_FLOWS;
	if (mlx5e_tc_int_port_supported(esw))
		nvports += MLX5E_TC_MAX_INT_PORT_NUM;

	return nvports;
}

static int esw_create_offloads_table(struct mlx5_eswitch *esw)
{
	struct mlx5_flow_table_attr ft_attr = {};
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_table *ft_offloads;
	struct mlx5_flow_namespace *ns;
	int err = 0;

	ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS);
	if (!ns) {
		esw_warn(esw->dev, "Failed to get offloads flow namespace\n");
		return -EOPNOTSUPP;
	}

	ft_attr.max_fte = esw_get_nr_ft_offloads_steering_src_ports(esw) +
			  MLX5_ESW_FT_OFFLOADS_DROP_RULE;
	ft_attr.prio = 1;

	ft_offloads = mlx5_create_flow_table(ns, &ft_attr);
	if (IS_ERR(ft_offloads)) {
		err = PTR_ERR(ft_offloads);
		esw_warn(esw->dev, "Failed to create offloads table, err %d\n", err);
		return err;
	}

	esw->offloads.ft_offloads = ft_offloads;
	return 0;
}

static void esw_destroy_offloads_table(struct mlx5_eswitch *esw)
{
	struct mlx5_esw_offload *offloads = &esw->offloads;

	mlx5_destroy_flow_table(offloads->ft_offloads);
}

static int esw_create_vport_rx_group(struct mlx5_eswitch *esw)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	u32 *flow_group_in;
	int nvports;
	int err = 0;

	nvports = esw_get_nr_ft_offloads_steering_src_ports(esw);
	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return -ENOMEM;

	mlx5_esw_set_flow_group_source_port(esw, flow_group_in, 0);

	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, nvports - 1);

	g = mlx5_create_flow_group(esw->offloads.ft_offloads, flow_group_in);

	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev, "Failed to create vport rx group err %d\n",
			 err);
		goto out;
	}

	esw->offloads.vport_rx_group = g;
out:
	kvfree(flow_group_in);
	return err;
}

static void esw_destroy_vport_rx_group(struct mlx5_eswitch *esw)
{
	mlx5_destroy_flow_group(esw->offloads.vport_rx_group);
}

static int esw_create_vport_rx_drop_rule_index(struct mlx5_eswitch *esw)
{
	/* ft_offloads table is enlarged by MLX5_ESW_FT_OFFLOADS_DROP_RULE (1)
	 * for the drop rule, which is placed at the end of the table.
	 * So return the total of vport and int_port as rule index.
	 */
	return esw_get_nr_ft_offloads_steering_src_ports(esw);
}

static int esw_create_vport_rx_drop_group(struct mlx5_eswitch *esw)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_group *g;
	u32 *flow_group_in;
	int flow_index;
	int err = 0;

	flow_index = esw_create_vport_rx_drop_rule_index(esw);

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return -ENOMEM;

	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index);

	g = mlx5_create_flow_group(esw->offloads.ft_offloads, flow_group_in);

	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(esw->dev,
			 "Failed to create vport rx drop group err %d\n", err);
		goto out;
	}

	esw->offloads.vport_rx_drop_group = g;
out:
	kvfree(flow_group_in);
	return err;
}

static void esw_destroy_vport_rx_drop_group(struct mlx5_eswitch *esw)
{
	if (esw->offloads.vport_rx_drop_group)
		mlx5_destroy_flow_group(esw->offloads.vport_rx_drop_group);
}

void
mlx5_esw_set_spec_source_port(struct mlx5_eswitch *esw,
			      u16 vport,
			      struct mlx5_flow_spec *spec)
{
	void *misc;

	if (mlx5_eswitch_vport_match_metadata_enabled(esw)) {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_for_match(esw, vport));

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters_2);
		MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
			 mlx5_eswitch_get_vport_metadata_mask());

		spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS_2;
	} else {
		misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters);
		MLX5_SET(fte_match_set_misc, misc, source_port, vport);

		misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
		MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);

		spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
	}
}

struct mlx5_flow_handle *
mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, u16 vport,
				  struct mlx5_flow_destination *dest)
{
	struct mlx5_flow_act flow_act = {0};
	struct mlx5_flow_handle *flow_rule;
	struct mlx5_flow_spec *spec;

	spec = kvzalloc_obj(*spec);
	if (!spec) {
		flow_rule = ERR_PTR(-ENOMEM);
		goto out;
	}

	mlx5_esw_set_spec_source_port(esw, vport, spec);

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	flow_rule = mlx5_add_flow_rules(esw->offloads.ft_offloads, spec,
					&flow_act, dest, 1);
	if (IS_ERR(flow_rule)) {
		esw_warn(esw->dev,
			 "fs offloads: Failed to add vport rx rule err %pe\n",
			 flow_rule);
		goto out;
	}

out:
	kvfree(spec);
	return flow_rule;
}

static int esw_create_vport_rx_drop_rule(struct mlx5_eswitch *esw)
{
	struct mlx5_flow_act flow_act = {};
	struct mlx5_flow_handle *flow_rule;

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP;
	flow_rule = mlx5_add_flow_rules(esw->offloads.ft_offloads, NULL,
					&flow_act, NULL, 0);
	if (IS_ERR(flow_rule)) {
		esw_warn(esw->dev,
			 "fs offloads: Failed to add vport rx drop rule err %pe\n",
			 flow_rule);
		return PTR_ERR(flow_rule);
	}

	esw->offloads.vport_rx_drop_rule = flow_rule;

	return 0;
}

static void esw_destroy_vport_rx_drop_rule(struct mlx5_eswitch *esw)
{
	if (esw->offloads.vport_rx_drop_rule)
		mlx5_del_flow_rules(esw->offloads.vport_rx_drop_rule);
}

static int mlx5_eswitch_inline_mode_get(struct mlx5_eswitch *esw, u8 *mode)
{
	u8 prev_mlx5_mode, mlx5_mode = MLX5_INLINE_MODE_L2;
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_vport *vport;
	unsigned long i;

	if (!MLX5_CAP_GEN(dev, vport_group_manager))
		return -EOPNOTSUPP;

	if (!mlx5_esw_is_fdb_created(esw))
		return -EOPNOTSUPP;

	switch (MLX5_CAP_ETH(dev, wqe_inline_mode)) {
	case MLX5_CAP_INLINE_MODE_NOT_REQUIRED:
		mlx5_mode = MLX5_INLINE_MODE_NONE;
		goto out;
	case MLX5_CAP_INLINE_MODE_L2:
		mlx5_mode = MLX5_INLINE_MODE_L2;
		goto out;
	case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT:
		goto query_vports;
	}

query_vports:
	mlx5_query_nic_vport_min_inline(dev, esw->first_host_vport, &prev_mlx5_mode);
	mlx5_esw_for_each_host_func_vport(esw, i, vport, esw->esw_funcs.num_vfs) {
		mlx5_query_nic_vport_min_inline(dev, vport->vport, &mlx5_mode);
		if (prev_mlx5_mode != mlx5_mode)
			return -EINVAL;
		prev_mlx5_mode = mlx5_mode;
	}

out:
	*mode = mlx5_mode;
	return 0;
}

static void esw_destroy_restore_table(struct mlx5_eswitch *esw)
{
	struct mlx5_esw_offload *offloads = &esw->offloads;

	if (!mlx5_eswitch_reg_c1_loopback_supported(esw))
		return;

	mlx5_modify_header_dealloc(esw->dev, offloads->restore_copy_hdr_id);
	mlx5_destroy_flow_group(offloads->restore_group);
	mlx5_destroy_flow_table(offloads->ft_offloads_restore);
}

static int esw_create_restore_table(struct mlx5_eswitch *esw)
{
	u8 modact[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {};
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_table_attr ft_attr = {};
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_namespace *ns;
	struct mlx5_modify_hdr *mod_hdr;
	void *match_criteria, *misc;
	struct mlx5_flow_table *ft;
	struct mlx5_flow_group *g;
	u32 *flow_group_in;
	int err = 0;

	if (!mlx5_eswitch_reg_c1_loopback_supported(esw))
		return 0;

	ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_OFFLOADS);
	if (!ns) {
		esw_warn(esw->dev, "Failed to get offloads flow namespace\n");
		return -EOPNOTSUPP;
	}

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in) {
		err = -ENOMEM;
		goto out_free;
	}

	ft_attr.max_fte = 1 << ESW_REG_C0_USER_DATA_METADATA_BITS;
	ft = mlx5_create_flow_table(ns, &ft_attr);
	if (IS_ERR(ft)) {
		err = PTR_ERR(ft);
		esw_warn(esw->dev, "Failed to create restore table, err %d\n",
			 err);
		goto out_free;
	}

	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in,
				      match_criteria);
	misc = MLX5_ADDR_OF(fte_match_param, match_criteria,
			    misc_parameters_2);

	MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_0,
		 ESW_REG_C0_USER_DATA_METADATA_MASK);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index,
		 ft_attr.max_fte - 1);
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_MISC_PARAMETERS_2);
	g = mlx5_create_flow_group(ft, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		esw_warn(dev, "Failed to create restore flow group, err: %d\n",
			 err);
		goto err_group;
	}

	MLX5_SET(copy_action_in, modact, action_type, MLX5_ACTION_TYPE_COPY);
	MLX5_SET(copy_action_in, modact, src_field,
		 MLX5_ACTION_IN_FIELD_METADATA_REG_C_1);
	MLX5_SET(copy_action_in, modact, dst_field,
		 MLX5_ACTION_IN_FIELD_METADATA_REG_B);
	mod_hdr = mlx5_modify_header_alloc(esw->dev,
					   MLX5_FLOW_NAMESPACE_KERNEL, 1,
					   modact);
	if (IS_ERR(mod_hdr)) {
		err = PTR_ERR(mod_hdr);
		esw_warn(dev, "Failed to create restore mod header, err: %d\n",
			 err);
		goto err_mod_hdr;
	}

	esw->offloads.ft_offloads_restore = ft;
	esw->offloads.restore_group = g;
	esw->offloads.restore_copy_hdr_id = mod_hdr;

	kvfree(flow_group_in);

	return 0;

err_mod_hdr:
	mlx5_destroy_flow_group(g);
err_group:
	mlx5_destroy_flow_table(ft);
out_free:
	kvfree(flow_group_in);

	return err;
}

static void mlx5_esw_assert_reps_locked(struct mlx5_eswitch *esw)
{
	lockdep_assert_held(&esw->offloads.reps_lock);
}

void mlx5_esw_reps_block(struct mlx5_eswitch *esw)
{
	mutex_lock(&esw->offloads.reps_lock);
}

static void mlx5_esw_reps_block_nested(struct mlx5_eswitch *esw)
{
	mutex_lock_nested(&esw->offloads.reps_lock, SINGLE_DEPTH_NESTING);
}

void mlx5_esw_reps_unblock(struct mlx5_eswitch *esw)
{
	mutex_unlock(&esw->offloads.reps_lock);
}

static void esw_mode_change(struct mlx5_eswitch *esw, u16 mode)
{
	mlx5_esw_reps_unblock(esw);
	mlx5_devcom_comp_lock(esw->dev->priv.hca_devcom_comp);
	if (esw->dev->priv.flags & MLX5_PRIV_FLAGS_DISABLE_IB_ADEV ||
	    mlx5_core_mp_enabled(esw->dev)) {
		esw->mode = mode;
		goto out;
	}

	esw->dev->priv.flags |= MLX5_PRIV_FLAGS_DISABLE_IB_ADEV;
	mlx5_rescan_drivers_locked(esw->dev);
	esw->mode = mode;
	esw->dev->priv.flags &= ~MLX5_PRIV_FLAGS_DISABLE_IB_ADEV;
out:
	mlx5_rescan_drivers_locked(esw->dev);
	mlx5_devcom_comp_unlock(esw->dev->priv.hca_devcom_comp);
	mlx5_esw_reps_block(esw);
}

static void mlx5_esw_fdb_drop_destroy(struct mlx5_eswitch *esw)
{
	if (!esw->fdb_table.offloads.drop_root)
		return;

	esw_debug(esw->dev, "Destroying FDB drop root table %#x fc %#x\n",
		  esw->fdb_table.offloads.drop_root->id,
		  esw->fdb_table.offloads.drop_root_fc->id);
	mlx5_del_flow_rules(esw->fdb_table.offloads.drop_root_rule);
	/* Don't free flow counter here, can be reused on a later activation */
	mlx5_destroy_flow_table(esw->fdb_table.offloads.drop_root);
	esw->fdb_table.offloads.drop_root_rule = NULL;
	esw->fdb_table.offloads.drop_root = NULL;
}

static int mlx5_esw_fdb_drop_create(struct mlx5_eswitch *esw)
{
	struct mlx5_flow_destination drop_fc_dst = {};
	struct mlx5_flow_table_attr ft_attr = {};
	struct mlx5_flow_destination *dst = NULL;
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_flow_namespace *root_ns;
	struct mlx5_flow_act flow_act = {};
	struct mlx5_flow_handle *flow_rule;
	struct mlx5_flow_table *table;
	int err = 0, dst_num = 0;

	if (esw->fdb_table.offloads.drop_root)
		return 0;

	root_ns = esw->fdb_table.offloads.ns;

	ft_attr.prio = FDB_DROP_ROOT;
	ft_attr.max_fte = 1;
	ft_attr.autogroup.max_num_groups = 1;
	table = mlx5_create_auto_grouped_flow_table(root_ns, &ft_attr);
	if (IS_ERR(table)) {
		esw_warn(dev, "Failed to create fdb drop root table, err %pe\n",
			 table);
		return PTR_ERR(table);
	}

	/* Drop FC reusable, create once on first deactivation of FDB */
	if (!esw->fdb_table.offloads.drop_root_fc) {
		struct mlx5_fc *counter = mlx5_fc_create(dev, 0);

		err = PTR_ERR_OR_ZERO(counter);
		if (err)
			esw_warn(esw->dev, "create fdb drop fc err %d\n", err);
		else
			esw->fdb_table.offloads.drop_root_fc = counter;
	}

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP;

	if (esw->fdb_table.offloads.drop_root_fc) {
		flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT;
		drop_fc_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER;
		drop_fc_dst.counter = esw->fdb_table.offloads.drop_root_fc;
		dst = &drop_fc_dst;
		dst_num++;
	}

	flow_rule = mlx5_add_flow_rules(table, NULL, &flow_act, dst, dst_num);
	err = PTR_ERR_OR_ZERO(flow_rule);
	if (err) {
		esw_warn(esw->dev,
			 "fs offloads: Failed to add vport rx drop rule err %d\n",
			 err);
		goto err_flow_rule;
	}

	esw->fdb_table.offloads.drop_root = table;
	esw->fdb_table.offloads.drop_root_rule = flow_rule;
	esw_debug(esw->dev, "Created FDB drop root table %#x fc %#x\n",
		  table->id, dst ? dst->counter->id : 0);
	return 0;

err_flow_rule:
	/* no need to free drop fc, esw_offloads_steering_cleanup will do it */
	mlx5_destroy_flow_table(table);
	return err;
}

static void mlx5_esw_fdb_active(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;
	unsigned long i;

	mlx5_esw_fdb_drop_destroy(esw);
	mlx5_mpfs_enable(esw->dev);

	mutex_lock(&esw->state_lock);
	mlx5_esw_for_each_vf_vport(esw, i, vport, U16_MAX) {
		if (!vport->adjacent)
			continue;
		esw_debug(esw->dev, "Connecting vport %d to eswitch\n",
			  vport->vport);
		mlx5_esw_adj_vport_modify(esw->dev, vport->vport, true);
	}
	mutex_unlock(&esw->state_lock);

	esw->offloads_inactive = false;
	esw_warn(esw->dev, "MPFS/FDB active\n");
}

static void mlx5_esw_fdb_inactive(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;
	unsigned long i;

	mlx5_mpfs_disable(esw->dev);
	mlx5_esw_fdb_drop_create(esw);

	mutex_lock(&esw->state_lock);
	mlx5_esw_for_each_vf_vport(esw, i, vport, U16_MAX) {
		if (!vport->adjacent)
			continue;
		esw_debug(esw->dev, "Disconnecting vport %u from eswitch\n",
			  vport->vport);

		mlx5_esw_adj_vport_modify(esw->dev, vport->vport, false);
	}
	mutex_unlock(&esw->state_lock);

	esw->offloads_inactive = true;
	esw_warn(esw->dev, "MPFS/FDB inactive\n");
}

static int esw_offloads_start(struct mlx5_eswitch *esw,
			      struct netlink_ext_ack *extack)
{
	int err;

	esw_mode_change(esw, MLX5_ESWITCH_OFFLOADS);
	err = mlx5_eswitch_enable_locked(esw, esw->dev->priv.sriov.num_vfs);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Failed setting eswitch to offloads");
		esw_mode_change(esw, MLX5_ESWITCH_LEGACY);
		return err;
	}
	if (esw->offloads.inline_mode == MLX5_INLINE_MODE_NONE) {
		if (mlx5_eswitch_inline_mode_get(esw,
						 &esw->offloads.inline_mode)) {
			esw->offloads.inline_mode = MLX5_INLINE_MODE_L2;
			NL_SET_ERR_MSG_MOD(extack,
					   "Inline mode is different between vports");
		}
	}
	return 0;
}

void mlx5_esw_offloads_rep_remove(struct mlx5_eswitch *esw,
				  const struct mlx5_vport *vport)
{
	struct mlx5_eswitch_rep *rep = xa_load(&esw->offloads.vport_reps,
					       vport->vport);

	if (!rep)
		return;
	xa_erase(&esw->offloads.vport_reps, vport->vport);
	kfree(rep);
}

int mlx5_esw_offloads_rep_add(struct mlx5_eswitch *esw,
			      const struct mlx5_vport *vport)
{
	struct mlx5_eswitch_rep *rep;
	int rep_type;
	int err;

	rep = kzalloc_obj(*rep);
	if (!rep)
		return -ENOMEM;

	rep->vport = vport->vport;
	rep->vport_index = vport->index;
	for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) {
		if (!esw->offloads.rep_ops[rep_type]) {
			atomic_set(&rep->rep_data[rep_type].state,
				   REP_UNREGISTERED);
			continue;
		}
		/* Dynamic/delegated vports add their representors after
		 * mlx5_eswitch_register_vport_reps, so mark them as registered
		 * for them to be loaded later with the others.
		 */
		rep->esw = esw;
		atomic_set(&rep->rep_data[rep_type].state, REP_REGISTERED);
	}
	err = xa_insert(&esw->offloads.vport_reps, rep->vport, rep, GFP_KERNEL);
	if (err)
		goto insert_err;

	return 0;

insert_err:
	kfree(rep);
	return err;
}

static void mlx5_esw_offloads_rep_cleanup(struct mlx5_eswitch *esw,
					  struct mlx5_eswitch_rep *rep)
{
	xa_erase(&esw->offloads.vport_reps, rep->vport);
	kfree(rep);
}

static void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long i;

	mlx5_esw_for_each_rep(esw, i, rep)
		mlx5_esw_offloads_rep_cleanup(esw, rep);
	xa_destroy(&esw->offloads.vport_reps);
	mutex_destroy(&esw->offloads.reps_lock);
}

static int esw_offloads_init_reps(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;
	unsigned long i;
	int err;

	xa_init(&esw->offloads.vport_reps);
	mutex_init(&esw->offloads.reps_lock);

	mlx5_esw_for_each_vport(esw, i, vport) {
		err = mlx5_esw_offloads_rep_add(esw, vport);
		if (err)
			goto err;
	}
	return 0;

err:
	esw_offloads_cleanup_reps(esw);
	return err;
}

static int esw_port_metadata_set(struct devlink *devlink, u32 id,
				 struct devlink_param_gset_ctx *ctx,
				 struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = devlink_priv(devlink);
	struct mlx5_eswitch *esw = dev->priv.eswitch;
	int err = 0;

	down_write(&esw->mode_lock);
	if (mlx5_esw_is_fdb_created(esw)) {
		err = -EBUSY;
		goto done;
	}
	if (!mlx5_esw_vport_match_metadata_supported(esw)) {
		err = -EOPNOTSUPP;
		goto done;
	}
	if (ctx->val.vbool)
		esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;
	else
		esw->flags &= ~MLX5_ESWITCH_VPORT_MATCH_METADATA;
done:
	up_write(&esw->mode_lock);
	return err;
}

static int esw_port_metadata_get(struct devlink *devlink, u32 id,
				 struct devlink_param_gset_ctx *ctx,
				 struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = devlink_priv(devlink);

	ctx->val.vbool = mlx5_eswitch_vport_match_metadata_enabled(dev->priv.eswitch);
	return 0;
}

static int esw_port_metadata_validate(struct devlink *devlink, u32 id,
				      union devlink_param_value *val,
				      struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = devlink_priv(devlink);
	u8 esw_mode;

	esw_mode = mlx5_eswitch_mode(dev);
	if (esw_mode == MLX5_ESWITCH_OFFLOADS) {
		NL_SET_ERR_MSG_MOD(extack,
				   "E-Switch must either disabled or non switchdev mode");
		return -EBUSY;
	}
	return 0;
}

static const struct devlink_param esw_devlink_params[] = {
	DEVLINK_PARAM_DRIVER(MLX5_DEVLINK_PARAM_ID_ESW_PORT_METADATA,
			     "esw_port_metadata", DEVLINK_PARAM_TYPE_BOOL,
			     BIT(DEVLINK_PARAM_CMODE_RUNTIME),
			     esw_port_metadata_get,
			     esw_port_metadata_set,
			     esw_port_metadata_validate),
};

int esw_offloads_init(struct mlx5_eswitch *esw)
{
	int err;

	err = esw_offloads_init_reps(esw);
	if (err)
		return err;

	if (MLX5_ESWITCH_MANAGER(esw->dev) &&
	    mlx5_esw_vport_match_metadata_supported(esw))
		esw->flags |= MLX5_ESWITCH_VPORT_MATCH_METADATA;

	err = devl_params_register(priv_to_devlink(esw->dev),
				   esw_devlink_params,
				   ARRAY_SIZE(esw_devlink_params));
	if (err)
		goto err_params;

	return 0;

err_params:
	esw_offloads_cleanup_reps(esw);
	return err;
}

void esw_offloads_cleanup(struct mlx5_eswitch *esw)
{
	devl_params_unregister(priv_to_devlink(esw->dev),
			       esw_devlink_params,
			       ARRAY_SIZE(esw_devlink_params));
	esw_offloads_cleanup_reps(esw);
}

static int __esw_offloads_load_rep(struct mlx5_eswitch *esw,
				   struct mlx5_eswitch_rep *rep,
				   u8 rep_type, bool *newly_loaded)
{
	int err;

	mlx5_esw_assert_reps_locked(esw);

	if (newly_loaded)
		*newly_loaded = false;

	if (atomic_cmpxchg(&rep->rep_data[rep_type].state,
			   REP_REGISTERED, REP_LOADED) != REP_REGISTERED)
		return 0;

	err = esw->offloads.rep_ops[rep_type]->load(esw->dev, rep);
	if (err) {
		atomic_set(&rep->rep_data[rep_type].state, REP_REGISTERED);
		return err;
	}

	if (newly_loaded)
		*newly_loaded = true;

	return 0;
}

static void __esw_offloads_unload_rep(struct mlx5_eswitch *esw,
				      struct mlx5_eswitch_rep *rep, u8 rep_type)
{
	mlx5_esw_assert_reps_locked(esw);

	if (atomic_cmpxchg(&rep->rep_data[rep_type].state,
			   REP_LOADED, REP_REGISTERED) == REP_LOADED) {
		if (rep_type == REP_ETH)
			__esw_offloads_unload_rep(esw, rep, REP_IB);
		esw->offloads.rep_ops[rep_type]->unload(rep);
	}
}

static void __unload_reps_all_vport(struct mlx5_eswitch *esw, u8 rep_type)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long i;

	mlx5_esw_for_each_rep(esw, i, rep)
		__esw_offloads_unload_rep(esw, rep, rep_type);
}

static int mlx5_esw_offloads_rep_load(struct mlx5_eswitch *esw, u16 vport_num)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long loaded = 0;
	bool newly_loaded;
	int rep_type;
	int err;

	if (vport_num != MLX5_VPORT_UPLINK &&
	    mlx5_get_sd(esw->dev) && !mlx5_lag_is_active(esw->dev))
		return 0;

	rep = mlx5_eswitch_get_rep(esw, vport_num);
	for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) {
		err = __esw_offloads_load_rep(esw, rep, rep_type,
					      &newly_loaded);
		if (err)
			goto err_reps;
		if (newly_loaded)
			loaded |= BIT(rep_type);
	}

	return 0;

err_reps:
	while (--rep_type >= 0)
		if (test_bit(rep_type, &loaded))
			__esw_offloads_unload_rep(esw, rep, rep_type);
	return err;
}

static void mlx5_esw_offloads_rep_unload(struct mlx5_eswitch *esw, u16 vport_num)
{
	struct mlx5_eswitch_rep *rep;
	int rep_type;

	rep = mlx5_eswitch_get_rep(esw, vport_num);
	for (rep_type = NUM_REP_TYPES - 1; rep_type >= 0; rep_type--)
		__esw_offloads_unload_rep(esw, rep, rep_type);
}

int mlx5_esw_offloads_init_pf_vf_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
{
	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return 0;

	return mlx5_esw_offloads_pf_vf_devlink_port_init(esw, vport);
}

void mlx5_esw_offloads_cleanup_pf_vf_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
{
	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return;

	mlx5_esw_offloads_pf_vf_devlink_port_cleanup(esw, vport);
}

int mlx5_esw_offloads_init_sf_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport,
				  struct mlx5_devlink_port *dl_port,
				  u32 controller, u32 sfnum)
{
	return mlx5_esw_offloads_sf_devlink_port_init(esw, vport, dl_port, controller, sfnum);
}

void mlx5_esw_offloads_cleanup_sf_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
{
	mlx5_esw_offloads_sf_devlink_port_cleanup(esw, vport);
}

int mlx5_esw_offloads_load_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
{
	int err;

	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return 0;

	err = mlx5_esw_offloads_devlink_port_register(esw, vport);
	if (err)
		return err;

	err = mlx5_esw_offloads_rep_load(esw, vport->vport);
	if (err)
		goto load_err;
	return err;

load_err:
	mlx5_esw_offloads_devlink_port_unregister(vport);
	return err;
}

void mlx5_esw_offloads_unload_rep(struct mlx5_eswitch *esw, struct mlx5_vport *vport)
{
	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return;

	mlx5_esw_offloads_rep_unload(esw, vport->vport);

	mlx5_esw_offloads_devlink_port_unregister(vport);
}

static int esw_set_slave_root_fdb(struct mlx5_core_dev *master,
				  struct mlx5_core_dev *slave)
{
	u32 in[MLX5_ST_SZ_DW(set_flow_table_root_in)]   = {};
	u32 out[MLX5_ST_SZ_DW(set_flow_table_root_out)] = {};
	struct mlx5_flow_root_namespace *root;
	struct mlx5_flow_namespace *ns;
	int err;

	MLX5_SET(set_flow_table_root_in, in, opcode,
		 MLX5_CMD_OP_SET_FLOW_TABLE_ROOT);
	MLX5_SET(set_flow_table_root_in, in, table_type,
		 FS_FT_FDB);

	if (master) {
		ns = mlx5_get_flow_namespace(master,
					     MLX5_FLOW_NAMESPACE_FDB);
		root = find_root(&ns->node);
		mutex_lock(&root->chain_lock);
		MLX5_SET(set_flow_table_root_in, in,
			 table_eswitch_owner_vhca_id_valid, 1);
		MLX5_SET(set_flow_table_root_in, in,
			 table_eswitch_owner_vhca_id,
			 MLX5_CAP_GEN(master, vhca_id));
		MLX5_SET(set_flow_table_root_in, in, table_id,
			 root->root_ft->id);
	} else {
		ns = mlx5_get_flow_namespace(slave,
					     MLX5_FLOW_NAMESPACE_FDB);
		root = find_root(&ns->node);
		mutex_lock(&root->chain_lock);
		MLX5_SET(set_flow_table_root_in, in, table_id,
			 root->root_ft->id);
	}

	err = mlx5_cmd_exec(slave, in, sizeof(in), out, sizeof(out));
	mutex_unlock(&root->chain_lock);

	return err;
}

static int __esw_set_master_egress_rule(struct mlx5_core_dev *master,
					struct mlx5_core_dev *slave,
					struct mlx5_vport *vport,
					struct mlx5_flow_table *acl)
{
	u16 slave_index = MLX5_CAP_GEN(slave, vhca_id);
	struct mlx5_flow_handle *flow_rule = NULL;
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_act flow_act = {};
	struct mlx5_flow_spec *spec;
	int err = 0;
	void *misc;

	spec = kvzalloc_obj(*spec);
	if (!spec)
		return -ENOMEM;

	spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS;
	misc = MLX5_ADDR_OF(fte_match_param, spec->match_value,
			    misc_parameters);
	MLX5_SET(fte_match_set_misc, misc, source_port, MLX5_VPORT_UPLINK);
	MLX5_SET(fte_match_set_misc, misc, source_eswitch_owner_vhca_id, slave_index);

	misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters);
	MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port);
	MLX5_SET_TO_ONES(fte_match_set_misc, misc,
			 source_eswitch_owner_vhca_id);

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest.vport.num = slave->priv.eswitch->manager_vport;
	dest.vport.vhca_id = MLX5_CAP_GEN(slave, vhca_id);
	dest.vport.flags |= MLX5_FLOW_DEST_VPORT_VHCA_ID;

	flow_rule = mlx5_add_flow_rules(acl, spec, &flow_act,
					&dest, 1);
	if (IS_ERR(flow_rule)) {
		err = PTR_ERR(flow_rule);
	} else {
		err = xa_insert(&vport->egress.offloads.bounce_rules,
				slave_index, flow_rule, GFP_KERNEL);
		if (err)
			mlx5_del_flow_rules(flow_rule);
	}

	kvfree(spec);
	return err;
}

static int esw_slave_egress_create_resources(struct mlx5_eswitch *esw,
					     struct mlx5_vport *vport)
{
	struct mlx5_flow_table_attr ft_attr = {
		.max_fte = 1, .prio = 0, .level = 0,
	};
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_namespace *ns;
	struct mlx5_flow_table *acl;
	struct mlx5_flow_group *g;
	u32 *flow_group_in;
	int err = 0;

	if (vport->egress.acl)
		return 0;

	xa_init_flags(&vport->egress.offloads.bounce_rules, XA_FLAGS_ALLOC);
	ns = mlx5_get_flow_vport_namespace(esw->dev,
					   MLX5_FLOW_NAMESPACE_ESW_EGRESS,
					   vport->index);
	if (!ns)
		return -EINVAL;

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return -ENOMEM;

	if (vport->vport || mlx5_core_is_ecpf(esw->dev))
		ft_attr.flags = MLX5_FLOW_TABLE_OTHER_VPORT;

	acl = mlx5_create_vport_flow_table(ns, &ft_attr, vport->vport);
	if (IS_ERR(acl)) {
		err = PTR_ERR(acl);
		goto out;
	}

	g = mlx5_create_flow_group(acl, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		goto err_table;
	}

	vport->egress.acl = acl;
	vport->egress.offloads.bounce_grp = g;
	vport->egress.type = VPORT_EGRESS_ACL_TYPE_SHARED_FDB;
	err = 0;

err_table:
	if (err && !IS_ERR_OR_NULL(acl)) {
		mlx5_destroy_flow_table(acl);
		vport->egress.acl = NULL;
	}
out:
	kvfree(flow_group_in);
	return err;
}

static void esw_slave_egress_destroy_resources(struct mlx5_vport *vport)
{
	if (!IS_ERR_OR_NULL(vport->egress.offloads.bounce_grp)) {
		mlx5_destroy_flow_group(vport->egress.offloads.bounce_grp);
		vport->egress.offloads.bounce_grp = NULL;
	}
	if (!IS_ERR_OR_NULL(vport->egress.acl)) {
		esw_acl_egress_ofld_cleanup(vport);
		xa_destroy(&vport->egress.offloads.bounce_rules);
	}
}

static int esw_set_slave_egress_rule(struct mlx5_core_dev *master,
				     struct mlx5_core_dev *slave)
{
	struct mlx5_eswitch *slave_esw = slave->priv.eswitch;
	u16 master_vhca = MLX5_CAP_GEN(master, vhca_id);
	struct mlx5_flow_destination dest = {};
	struct mlx5_flow_handle *bounce_rule;
	struct mlx5_flow_act flow_act = {};
	struct mlx5_vport *slave_vport;
	int err;

	slave_vport = mlx5_eswitch_get_vport(slave_esw,
					     slave_esw->manager_vport);
	if (IS_ERR(slave_vport))
		return PTR_ERR(slave_vport);

	err = esw_slave_egress_create_resources(slave_esw, slave_vport);
	if (err)
		return err;

	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
	dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT;
	dest.vport.num = master->priv.eswitch->manager_vport;
	dest.vport.vhca_id = master_vhca;
	dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;

	bounce_rule = mlx5_add_flow_rules(slave_vport->egress.acl, NULL,
					  &flow_act, &dest, 1);
	if (IS_ERR(bounce_rule)) {
		err = PTR_ERR(bounce_rule);
		goto err_rule;
	}
	err = xa_insert(&slave_vport->egress.offloads.bounce_rules,
			master_vhca, bounce_rule, GFP_KERNEL);
	if (err)
		goto err_insert;

	return 0;
err_insert:
	mlx5_del_flow_rules(bounce_rule);
err_rule:
	esw_slave_egress_destroy_resources(slave_vport);
	return err;
}

static void esw_unset_slave_egress_rule(struct mlx5_core_dev *master,
					struct mlx5_core_dev *slave)
{
	struct mlx5_eswitch *slave_esw = slave->priv.eswitch;
	u16 master_vhca = MLX5_CAP_GEN(master, vhca_id);
	struct mlx5_vport *slave_vport;

	slave_vport = mlx5_eswitch_get_vport(slave_esw,
					     slave_esw->manager_vport);
	if (IS_ERR(slave_vport))
		return;

	esw_acl_egress_ofld_bounce_rule_destroy(slave_vport, master_vhca);
	esw_slave_egress_destroy_resources(slave_vport);
}

static int esw_master_egress_create_resources(struct mlx5_eswitch *esw,
					      struct mlx5_flow_namespace *egress_ns,
					      struct mlx5_vport *vport, size_t count)
{
	int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in);
	struct mlx5_flow_table_attr ft_attr = {
		.max_fte = count, .prio = 0, .level = 0,
	};
	struct mlx5_flow_table *acl;
	struct mlx5_flow_group *g;
	void *match_criteria;
	u32 *flow_group_in;
	int err;

	if (vport->egress.acl)
		return 0;

	flow_group_in = kvzalloc(inlen, GFP_KERNEL);
	if (!flow_group_in)
		return -ENOMEM;

	if (vport->vport || mlx5_core_is_ecpf(esw->dev))
		ft_attr.flags = MLX5_FLOW_TABLE_OTHER_VPORT;

	acl = mlx5_create_vport_flow_table(egress_ns, &ft_attr, vport->vport);
	if (IS_ERR(acl)) {
		err = PTR_ERR(acl);
		goto out;
	}

	match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in,
				      match_criteria);
	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
			 misc_parameters.source_port);
	MLX5_SET_TO_ONES(fte_match_param, match_criteria,
			 misc_parameters.source_eswitch_owner_vhca_id);
	MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable,
		 MLX5_MATCH_MISC_PARAMETERS);

	MLX5_SET(create_flow_group_in, flow_group_in,
		 source_eswitch_owner_vhca_id_valid, 1);
	MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0);
	MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, count);

	g = mlx5_create_flow_group(acl, flow_group_in);
	if (IS_ERR(g)) {
		err = PTR_ERR(g);
		goto err_group;
	}

	vport->egress.acl = acl;
	vport->egress.offloads.bounce_grp = g;
	vport->egress.type = VPORT_EGRESS_ACL_TYPE_SHARED_FDB;
	xa_init_flags(&vport->egress.offloads.bounce_rules, XA_FLAGS_ALLOC);

	kvfree(flow_group_in);

	return 0;

err_group:
	mlx5_destroy_flow_table(acl);
out:
	kvfree(flow_group_in);
	return err;
}

static void esw_master_egress_destroy_resources(struct mlx5_vport *vport)
{
	if (!xa_empty(&vport->egress.offloads.bounce_rules))
		return;
	mlx5_destroy_flow_group(vport->egress.offloads.bounce_grp);
	vport->egress.offloads.bounce_grp = NULL;
	mlx5_destroy_flow_table(vport->egress.acl);
	vport->egress.acl = NULL;
}

static int esw_set_master_egress_rule(struct mlx5_core_dev *master,
				      struct mlx5_core_dev *slave, size_t count)
{
	struct mlx5_eswitch *esw = master->priv.eswitch;
	u16 slave_index = MLX5_CAP_GEN(slave, vhca_id);
	struct mlx5_flow_namespace *egress_ns;
	struct mlx5_vport *vport;
	int err;

	vport = mlx5_eswitch_get_vport(esw, esw->manager_vport);
	if (IS_ERR(vport))
		return PTR_ERR(vport);

	egress_ns = mlx5_get_flow_vport_namespace(master,
						  MLX5_FLOW_NAMESPACE_ESW_EGRESS,
						  vport->index);
	if (!egress_ns)
		return -EINVAL;

	if (vport->egress.acl && vport->egress.type != VPORT_EGRESS_ACL_TYPE_SHARED_FDB)
		return 0;

	err = esw_master_egress_create_resources(esw, egress_ns, vport, count);
	if (err)
		return err;

	if (xa_load(&vport->egress.offloads.bounce_rules, slave_index))
		return -EINVAL;

	err = __esw_set_master_egress_rule(master, slave, vport, vport->egress.acl);
	if (err)
		goto err_rule;

	return 0;

err_rule:
	esw_master_egress_destroy_resources(vport);
	return err;
}

static void esw_unset_master_egress_rule(struct mlx5_core_dev *dev,
					 struct mlx5_core_dev *slave_dev)
{
	struct mlx5_vport *vport;

	vport = mlx5_eswitch_get_vport(dev->priv.eswitch,
				       dev->priv.eswitch->manager_vport);

	if (!vport->egress.acl)
		return;

	esw_acl_egress_ofld_bounce_rule_destroy(vport, MLX5_CAP_GEN(slave_dev, vhca_id));

	if (xa_empty(&vport->egress.offloads.bounce_rules)) {
		esw_acl_egress_ofld_cleanup(vport);
		xa_destroy(&vport->egress.offloads.bounce_rules);
	}
}

int mlx5_eswitch_offloads_single_fdb_add_one(struct mlx5_eswitch *master_esw,
					     struct mlx5_eswitch *slave_esw, int max_slaves)
{
	int err;

	err = esw_set_slave_root_fdb(master_esw->dev,
				     slave_esw->dev);
	if (err)
		return err;

	if (!mlx5_sd_is_primary(slave_esw->dev))
		return 0;

	err = esw_set_master_egress_rule(master_esw->dev,
					 slave_esw->dev, max_slaves);
	if (err)
		goto err_acl;

	return err;

err_acl:
	esw_set_slave_root_fdb(NULL, slave_esw->dev);
	return err;
}

void mlx5_eswitch_offloads_single_fdb_del_one(struct mlx5_eswitch *master_esw,
					      struct mlx5_eswitch *slave_esw)
{
	esw_set_slave_root_fdb(NULL, slave_esw->dev);
	esw_unset_master_egress_rule(master_esw->dev, slave_esw->dev);
}

int mlx5_eswitch_offloads_vport_lag_add_one(struct mlx5_eswitch *master_esw,
					    struct mlx5_eswitch *slave_esw)
{
	return esw_set_slave_egress_rule(master_esw->dev, slave_esw->dev);
}

void mlx5_eswitch_offloads_vport_lag_del_one(struct mlx5_eswitch *master_esw,
					     struct mlx5_eswitch *slave_esw)
{
	esw_unset_slave_egress_rule(master_esw->dev, slave_esw->dev);
}

#define ESW_OFFLOADS_DEVCOM_PAIR	(0)
#define ESW_OFFLOADS_DEVCOM_UNPAIR	(1)

static void mlx5_esw_offloads_rep_event_unpair(struct mlx5_eswitch *esw,
					       struct mlx5_eswitch *peer_esw)
{
	const struct mlx5_eswitch_rep_ops *ops;
	struct mlx5_eswitch_rep *rep;
	unsigned long i;
	u8 rep_type;

	mlx5_esw_for_each_rep(esw, i, rep) {
		rep_type = NUM_REP_TYPES;
		while (rep_type--) {
			ops = esw->offloads.rep_ops[rep_type];
			if (atomic_read(&rep->rep_data[rep_type].state) == REP_LOADED &&
			    ops->event)
				ops->event(esw, rep, MLX5_SWITCHDEV_EVENT_UNPAIR, peer_esw);
		}
	}
}

static void mlx5_esw_offloads_unpair(struct mlx5_eswitch *esw,
				     struct mlx5_eswitch *peer_esw)
{
#if IS_ENABLED(CONFIG_MLX5_CLS_ACT)
	mlx5e_tc_clean_fdb_peer_flows(esw);
#endif
	mlx5_esw_offloads_rep_event_unpair(esw, peer_esw);
	esw_del_fdb_peer_miss_rules(esw, peer_esw->dev);
}

static int mlx5_esw_offloads_pair(struct mlx5_eswitch *esw,
				  struct mlx5_eswitch *peer_esw)
{
	const struct mlx5_eswitch_rep_ops *ops;
	struct mlx5_eswitch_rep *rep;
	unsigned long i;
	u8 rep_type;
	int err;

	err = esw_add_fdb_peer_miss_rules(esw, peer_esw->dev);
	if (err)
		return err;

	mlx5_esw_for_each_rep(esw, i, rep) {
		for (rep_type = 0; rep_type < NUM_REP_TYPES; rep_type++) {
			ops = esw->offloads.rep_ops[rep_type];
			if (atomic_read(&rep->rep_data[rep_type].state) == REP_LOADED &&
			    ops->event) {
				err = ops->event(esw, rep, MLX5_SWITCHDEV_EVENT_PAIR, peer_esw);
				if (err)
					goto err_out;
			}
		}
	}

	return 0;

err_out:
	mlx5_esw_offloads_unpair(esw, peer_esw);
	return err;
}

static int mlx5_esw_offloads_set_ns_peer(struct mlx5_eswitch *esw,
					 struct mlx5_eswitch *peer_esw,
					 bool pair)
{
	u16 peer_vhca_id = MLX5_CAP_GEN(peer_esw->dev, vhca_id);
	u16 vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id);
	struct mlx5_flow_root_namespace *peer_ns;
	struct mlx5_flow_root_namespace *ns;
	int err;

	peer_ns = peer_esw->dev->priv.steering->fdb_root_ns;
	ns = esw->dev->priv.steering->fdb_root_ns;

	if (pair) {
		err = mlx5_flow_namespace_set_peer(ns, peer_ns, peer_vhca_id);
		if (err)
			return err;

		err = mlx5_flow_namespace_set_peer(peer_ns, ns, vhca_id);
		if (err) {
			mlx5_flow_namespace_set_peer(ns, NULL, peer_vhca_id);
			return err;
		}
	} else {
		mlx5_flow_namespace_set_peer(ns, NULL, peer_vhca_id);
		mlx5_flow_namespace_set_peer(peer_ns, NULL, vhca_id);
	}

	return 0;
}

bool mlx5_eswitch_is_peer(struct mlx5_eswitch *esw,
			  struct mlx5_eswitch *peer_esw)
{
	u16 peer_esw_i;

	if (!mlx5_esw_allowed(esw) || !mlx5_esw_allowed(peer_esw))
		return false;

	peer_esw_i = MLX5_CAP_GEN(peer_esw->dev, vhca_id);
	return !!xa_load(&esw->paired, peer_esw_i);
}

static int mlx5_esw_offloads_devcom_event(int event,
					  void *my_data,
					  void *event_data)
{
	struct mlx5_eswitch *esw = my_data;
	struct mlx5_eswitch *peer_esw = event_data;
	u16 esw_i, peer_esw_i;
	bool esw_paired;
	int err;

	peer_esw_i = MLX5_CAP_GEN(peer_esw->dev, vhca_id);
	esw_i = MLX5_CAP_GEN(esw->dev, vhca_id);
	esw_paired = !!xa_load(&esw->paired, peer_esw_i);

	switch (event) {
	case ESW_OFFLOADS_DEVCOM_PAIR:
		if (mlx5_eswitch_vport_match_metadata_enabled(esw) !=
		    mlx5_eswitch_vport_match_metadata_enabled(peer_esw))
			break;

		if (esw_paired)
			break;

		err = mlx5_esw_offloads_set_ns_peer(esw, peer_esw, true);
		if (err)
			goto err_out;

		err = mlx5_esw_offloads_pair(esw, peer_esw);
		if (err)
			goto err_peer;

		err = mlx5_esw_offloads_pair(peer_esw, esw);
		if (err)
			goto err_pair;

		err = xa_insert(&esw->paired, peer_esw_i, peer_esw, GFP_KERNEL);
		if (err)
			goto err_xa;

		err = xa_insert(&peer_esw->paired, esw_i, esw, GFP_KERNEL);
		if (err)
			goto err_peer_xa;

		esw->num_peers++;
		peer_esw->num_peers++;
		mlx5_devcom_comp_set_ready(esw->devcom, true);
		break;

	case ESW_OFFLOADS_DEVCOM_UNPAIR:
		if (!esw_paired)
			break;

		peer_esw->num_peers--;
		esw->num_peers--;
		if (!esw->num_peers && !peer_esw->num_peers)
			mlx5_devcom_comp_set_ready(esw->devcom, false);
		xa_erase(&peer_esw->paired, esw_i);
		xa_erase(&esw->paired, peer_esw_i);
		mlx5_esw_offloads_unpair(peer_esw, esw);
		mlx5_esw_offloads_unpair(esw, peer_esw);
		mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
		break;
	}

	return 0;

err_peer_xa:
	xa_erase(&esw->paired, peer_esw_i);
err_xa:
	mlx5_esw_offloads_unpair(peer_esw, esw);
err_pair:
	mlx5_esw_offloads_unpair(esw, peer_esw);
err_peer:
	mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
err_out:
	mlx5_core_err(esw->dev, "esw offloads devcom event failure, event %u err %d",
		      event, err);
	return err;
}

void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw,
				   const struct mlx5_devcom_match_attr *attr)
{
	int i;

	for (i = 0; i < MLX5_MAX_PORTS; i++)
		INIT_LIST_HEAD(&esw->offloads.peer_flows[i]);
	mutex_init(&esw->offloads.peer_mutex);

	if (!MLX5_CAP_ESW(esw->dev, merged_eswitch))
		return;

	if ((MLX5_VPORT_MANAGER(esw->dev) || mlx5_core_is_ecpf_esw_manager(esw->dev)) &&
	    (!mlx5_lag_is_supported(esw->dev) && !mlx5_get_sd(esw->dev)))
		return;

	xa_init(&esw->paired);
	xa_init(&esw->fdb_table.offloads.peer_miss_rules);
	esw->num_peers = 0;
	esw->devcom = mlx5_devcom_register_component(esw->dev->priv.devc,
						     MLX5_DEVCOM_ESW_OFFLOADS,
						     attr,
						     mlx5_esw_offloads_devcom_event,
						     esw);
	if (!esw->devcom)
		return;

	mlx5_devcom_send_event(esw->devcom,
			       ESW_OFFLOADS_DEVCOM_PAIR,
			       ESW_OFFLOADS_DEVCOM_UNPAIR,
			       esw);
}

void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw)
{
	if (!esw->devcom)
		return;

	mlx5_devcom_send_event(esw->devcom,
			       ESW_OFFLOADS_DEVCOM_UNPAIR,
			       ESW_OFFLOADS_DEVCOM_UNPAIR,
			       esw);

	mlx5_devcom_unregister_component(esw->devcom);
	xa_destroy(&esw->paired);
	xa_destroy(&esw->fdb_table.offloads.peer_miss_rules);
	esw->devcom = NULL;
}

bool mlx5_esw_offloads_devcom_is_ready(struct mlx5_eswitch *esw)
{
	return mlx5_devcom_comp_is_ready(esw->devcom);
}

bool mlx5_esw_vport_match_metadata_supported(const struct mlx5_eswitch *esw)
{
	if (!MLX5_CAP_ESW(esw->dev, esw_uplink_ingress_acl))
		return false;

	if (!(MLX5_CAP_ESW_FLOWTABLE(esw->dev, fdb_to_vport_reg_c_id) &
	      MLX5_FDB_TO_VPORT_REG_C_0))
		return false;

	return true;
}

#define MLX5_ESW_METADATA_RSVD_UPLINK 1

/* Share the same metadata for uplink's. This is fine because:
 * (a) In shared FDB mode (LAG) both uplink's are treated the
 *     same and tagged with the same metadata.
 * (b) In non shared FDB mode, packets from physical port0
 *     cannot hit eswitch of PF1 and vice versa.
 */
static u32 mlx5_esw_match_metadata_reserved(struct mlx5_eswitch *esw)
{
	return MLX5_ESW_METADATA_RSVD_UPLINK;
}

u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw)
{
	u32 vport_end_ida = (1 << ESW_VPORT_BITS) - 1;
	/* Reserve 0xf for internal port offload */
	u32 max_pf_num = (1 << ESW_PFNUM_BITS) - 2;
	int pf_num;
	int id;

	/* Only 4 bits of pf_num */
	pf_num = mlx5_sd_pf_num_get(esw->dev);
	if (pf_num < 0 || pf_num > max_pf_num)
		return 0;

	/* Metadata is 4 bits of PFNUM and 12 bits of unique id */
	/* Use only non-zero vport_id (2-4095) for all PF's */
	id = ida_alloc_range(&esw->offloads.vport_metadata_ida,
			     MLX5_ESW_METADATA_RSVD_UPLINK + 1,
			     vport_end_ida, GFP_KERNEL);
	if (id < 0)
		return 0;
	id = (pf_num << ESW_VPORT_BITS) | id;
	return id;
}

void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata)
{
	u32 vport_bit_mask = (1 << ESW_VPORT_BITS) - 1;

	/* Metadata contains only 12 bits of actual ida id */
	ida_free(&esw->offloads.vport_metadata_ida, metadata & vport_bit_mask);
}

static int esw_offloads_vport_metadata_setup(struct mlx5_eswitch *esw,
					     struct mlx5_vport *vport)
{
	if (vport->vport == MLX5_VPORT_UPLINK)
		vport->default_metadata = mlx5_esw_match_metadata_reserved(esw);
	else
		vport->default_metadata = mlx5_esw_match_metadata_alloc(esw);

	vport->metadata = vport->default_metadata;
	return vport->metadata ? 0 : -ENOSPC;
}

static void esw_offloads_vport_metadata_cleanup(struct mlx5_eswitch *esw,
						struct mlx5_vport *vport)
{
	if (!vport->default_metadata)
		return;

	if (vport->vport == MLX5_VPORT_UPLINK)
		return;

	WARN_ON(vport->metadata != vport->default_metadata);
	mlx5_esw_match_metadata_free(esw, vport->default_metadata);
	vport->default_metadata = 0;
}

static void esw_offloads_metadata_uninit(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;
	unsigned long i;

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw))
		return;

	mlx5_esw_for_each_vport(esw, i, vport)
		esw_offloads_vport_metadata_cleanup(esw, vport);
}

static int esw_offloads_metadata_init(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;
	unsigned long i;
	int err;

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw))
		return 0;

	mlx5_esw_for_each_vport(esw, i, vport) {
		err = esw_offloads_vport_metadata_setup(esw, vport);
		if (err)
			goto metadata_err;
	}

	return 0;

metadata_err:
	esw_offloads_metadata_uninit(esw);
	return err;
}

/* Deferred metadata init for SD devices: allocate vport metadata and
 * refresh the ingress ACL for every vport whose ACL was created with
 * metadata=0 in esw_create_offloads_acl_tables() / esw_vport_setup().
 *
 * No Rep is loaded at this point ==> no Rep net-dev exists, so no need
 * to take rtnl lock.
 *
 * Safe to call multiple times - subsequent calls are no-ops.
 */
int mlx5_esw_offloads_init_deferred_metadata(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *manager, *vport;
	unsigned long i;
	int err;

	if (!mlx5_eswitch_vport_match_metadata_enabled(esw))
		return 0;

	manager = mlx5_eswitch_get_vport(esw, esw->manager_vport);
	if (IS_ERR(manager))
		return PTR_ERR(manager);

	/* Sanity check: skip if metadata was already initialized */
	if (manager->default_metadata)
		return 0;

	err = esw_offloads_metadata_init(esw);
	if (err)
		return err;

	mutex_lock(&esw->state_lock);
	/* Manager vport doesn't have a rep/netdev loaded but its ingress ACL
	 * was programmed with metadata=0 - refresh it explicitly.
	 */
	err = mlx5_esw_acl_ingress_vport_metadata_update(esw,
							 esw->manager_vport,
							 0);
	if (err)
		goto err_acl;

	/* UPLINK is never marked enabled but its ACL is programmed in
	 * esw_create_offloads_acl_tables(); refresh it explicitly.
	 */
	err = mlx5_esw_acl_ingress_vport_metadata_update(esw, MLX5_VPORT_UPLINK,
							 0);
	if (err)
		goto err_acl;

	mlx5_esw_for_each_vport(esw, i, vport) {
		if (!vport || !vport->enabled)
			continue;
		err = mlx5_esw_acl_ingress_vport_metadata_update(esw,
								 vport->vport,
								 0);
		if (err)
			goto err_acl;
	}

	mutex_unlock(&esw->state_lock);
	return 0;

err_acl:
	esw_offloads_metadata_uninit(esw);
	mutex_unlock(&esw->state_lock);
	return err;
}

int
esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw,
				     struct mlx5_vport *vport)
{
	int err;

	err = esw_acl_ingress_ofld_setup(esw, vport);
	if (err)
		return err;

	err = esw_acl_egress_ofld_setup(esw, vport);
	if (err)
		goto egress_err;

	return 0;

egress_err:
	esw_acl_ingress_ofld_cleanup(esw, vport);
	return err;
}

void
esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw,
				      struct mlx5_vport *vport)
{
	esw_acl_egress_ofld_cleanup(vport);
	esw_acl_ingress_ofld_cleanup(esw, vport);
}

static int esw_create_offloads_acl_tables(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *uplink, *manager;
	int ret;

	uplink = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK);
	if (IS_ERR(uplink))
		return PTR_ERR(uplink);

	ret = esw_vport_create_offloads_acl_tables(esw, uplink);
	if (ret)
		return ret;

	manager = mlx5_eswitch_get_vport(esw, esw->manager_vport);
	if (IS_ERR(manager)) {
		ret = PTR_ERR(manager);
		goto err_manager;
	}

	ret = esw_vport_create_offloads_acl_tables(esw, manager);
	if (ret)
		goto err_manager;

	return 0;

err_manager:
	esw_vport_destroy_offloads_acl_tables(esw, uplink);
	return ret;
}

static void esw_destroy_offloads_acl_tables(struct mlx5_eswitch *esw)
{
	struct mlx5_vport *vport;

	vport = mlx5_eswitch_get_vport(esw, esw->manager_vport);
	if (!IS_ERR(vport))
		esw_vport_destroy_offloads_acl_tables(esw, vport);

	vport = mlx5_eswitch_get_vport(esw, MLX5_VPORT_UPLINK);
	if (!IS_ERR(vport))
		esw_vport_destroy_offloads_acl_tables(esw, vport);
}

void mlx5_eswitch_unload_reps(struct mlx5_eswitch *esw)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long i;

	if (!esw || esw->mode != MLX5_ESWITCH_OFFLOADS)
		return;

	mlx5_esw_for_each_rep(esw, i, rep) {
		if (rep->vport == MLX5_VPORT_UPLINK)
			continue;
		mlx5_esw_offloads_rep_unload(esw, rep->vport);
	}
}

int mlx5_eswitch_reload_ib_reps(struct mlx5_eswitch *esw)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long i;
	int ret;

	if (!esw || esw->mode != MLX5_ESWITCH_OFFLOADS)
		return 0;

	rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
	if (atomic_read(&rep->rep_data[REP_ETH].state) != REP_LOADED)
		return 0;

	/* SD secondary devices share the primary's uplink and do not
	 * have their own uplink representor. Only load VF/SF vports.
	 */
	if (mlx5_sd_is_primary(esw->dev)) {
		ret = __esw_offloads_load_rep(esw, rep, REP_IB, NULL);
		if (ret)
			return ret;
	}

	mlx5_esw_for_each_rep(esw, i, rep) {
		if (!mlx5_sd_is_primary(esw->dev) &&
		    rep->vport == MLX5_VPORT_UPLINK)
			continue;
		if (rep->vport != MLX5_VPORT_UPLINK &&
		    mlx5_get_sd(esw->dev) && !mlx5_lag_is_active(esw->dev))
			continue;

		if (atomic_read(&rep->rep_data[REP_ETH].state) == REP_LOADED)
			__esw_offloads_load_rep(esw, rep, REP_IB, NULL);
	}

	return 0;
}

static int esw_offloads_steering_init(struct mlx5_eswitch *esw)
{
	struct mlx5_esw_indir_table *indir;
	int err;

	memset(&esw->fdb_table.offloads, 0, sizeof(struct offloads_fdb));
	mutex_init(&esw->fdb_table.offloads.vports.lock);
	hash_init(esw->fdb_table.offloads.vports.table);
	atomic64_set(&esw->user_count, 0);

	indir = mlx5_esw_indir_table_init();
	if (IS_ERR(indir)) {
		err = PTR_ERR(indir);
		goto create_indir_err;
	}
	esw->fdb_table.offloads.indir = indir;

	err = esw_create_offloads_acl_tables(esw);
	if (err)
		goto create_acl_err;

	err = esw_create_offloads_table(esw);
	if (err)
		goto create_offloads_err;

	err = esw_create_restore_table(esw);
	if (err)
		goto create_restore_err;

	err = esw_create_offloads_fdb_tables(esw);
	if (err)
		goto create_fdb_err;

	err = esw_create_vport_rx_group(esw);
	if (err)
		goto create_fg_err;

	err = esw_create_vport_rx_drop_group(esw);
	if (err)
		goto create_rx_drop_fg_err;

	err = esw_create_vport_rx_drop_rule(esw);
	if (err)
		goto create_rx_drop_rule_err;

	return 0;

create_rx_drop_rule_err:
	esw_destroy_vport_rx_drop_group(esw);
create_rx_drop_fg_err:
	esw_destroy_vport_rx_group(esw);
create_fg_err:
	esw_destroy_offloads_fdb_tables(esw);
create_fdb_err:
	esw_destroy_restore_table(esw);
create_restore_err:
	esw_destroy_offloads_table(esw);
create_offloads_err:
	esw_destroy_offloads_acl_tables(esw);
create_acl_err:
	mlx5_esw_indir_table_destroy(esw->fdb_table.offloads.indir);
create_indir_err:
	mutex_destroy(&esw->fdb_table.offloads.vports.lock);
	return err;
}

static void esw_offloads_steering_cleanup(struct mlx5_eswitch *esw)
{
	mlx5_esw_fdb_drop_destroy(esw);
	if (esw->fdb_table.offloads.drop_root_fc)
		mlx5_fc_destroy(esw->dev, esw->fdb_table.offloads.drop_root_fc);
	esw->fdb_table.offloads.drop_root_fc = NULL;
	esw_destroy_vport_rx_drop_rule(esw);
	esw_destroy_vport_rx_drop_group(esw);
	esw_destroy_vport_rx_group(esw);
	esw_destroy_offloads_fdb_tables(esw);
	esw_destroy_restore_table(esw);
	esw_destroy_offloads_table(esw);
	esw_destroy_offloads_acl_tables(esw);
	mlx5_esw_indir_table_destroy(esw->fdb_table.offloads.indir);
	mutex_destroy(&esw->fdb_table.offloads.vports.lock);
}

static void esw_changed_event_handler(struct mlx5_eswitch *esw)
{
	struct mlx5_esw_pf_info host_pf_info;
	u16 new_num_vfs;
	const u32 *out;

	out = mlx5_esw_query_functions(esw->dev);
	if (IS_ERR(out))
		return;

	host_pf_info = mlx5_esw_get_host_pf_info(esw->dev, out);
	new_num_vfs = host_pf_info.num_of_vfs;

	if (host_pf_info.pf_disabled) {
		mlx5_sf_table_esw_changed_event_handler(esw->dev);
		mlx5_sf_hw_table_esw_changed_event_handler(esw->dev);
	}

	if (new_num_vfs == esw->esw_funcs.num_vfs || host_pf_info.pf_disabled)
		goto free;

	mlx5_esw_reps_block(esw);
	/* Number of VFs can only change from "0 to x" or "x to 0". */
	if (esw->esw_funcs.num_vfs > 0) {
		mlx5_eswitch_unload_vf_vports(esw, esw->esw_funcs.num_vfs);
	} else {
		int err;

		err = mlx5_eswitch_load_vf_vports(esw, new_num_vfs,
						  MLX5_VPORT_UC_ADDR_CHANGE);
		if (err)
			goto unblock;
	}
	esw->esw_funcs.num_vfs = new_num_vfs;
unblock:
	mlx5_esw_reps_unblock(esw);
free:
	kvfree(out);
}

static void esw_wq_handler(struct work_struct *work)
{
	struct mlx5_host_work *host_work;
	struct mlx5_eswitch *esw;
	struct devlink *devlink;
	int work_gen;

	host_work = container_of(work, struct mlx5_host_work, work);
	esw = host_work->esw;
	work_gen = host_work->work_gen;
	devlink = priv_to_devlink(esw->dev);

	/* Do not block on devlink lock until stale work is filtered out.
	 * Teardown can invalidate the generation and then wait for this
	 * workqueue while holding devlink lock.
	 */
	for (;;) {
		if (work_gen != atomic_read(&esw->generation))
			goto free;

		if (devl_trylock(devlink))
			break;

		wait_event_timeout(esw->work_queue_wait,
				   work_gen != atomic_read(&esw->generation),
				   msecs_to_jiffies(60));
	}

	/* Stale work from one or more mode changes ago. Bail out. */
	if (work_gen != atomic_read(&esw->generation))
		goto unlock;

	host_work->func(esw);

unlock:
	devl_unlock(devlink);
free:
	kfree(host_work);
}

static int mlx5_esw_add_work(struct mlx5_eswitch *esw,
			     void (*func)(struct mlx5_eswitch *esw),
			     gfp_t gfp)
{
	struct mlx5_host_work *host_work;

	host_work = kzalloc_obj(*host_work, gfp);
	if (!host_work)
		return -ENOMEM;

	host_work->esw = esw;
	host_work->work_gen = atomic_read(&esw->generation);

	host_work->func = func;
	INIT_WORK(&host_work->work, esw_wq_handler);
	queue_work(esw->work_queue, &host_work->work);

	return 0;
}

int mlx5_esw_funcs_changed_handler(struct notifier_block *nb,
				   unsigned long type, void *data)
{
	struct mlx5_esw_functions *esw_funcs;
	struct mlx5_eswitch *esw;
	int ret;

	esw_funcs = mlx5_nb_cof(nb, struct mlx5_esw_functions, nb);
	esw = container_of(esw_funcs, struct mlx5_eswitch, esw_funcs);

	ret = mlx5_esw_add_work(esw, esw_changed_event_handler, GFP_ATOMIC);
	if (ret)
		return NOTIFY_DONE;

	return NOTIFY_OK;
}

bool mlx5_esw_offloads_controller_valid(const struct mlx5_eswitch *esw, u32 controller)
{
	const struct mlx5_esw_functions *esw_funcs;
	int i;

	/* Local controller is always valid */
	if (controller == 0)
		return true;

	if (!mlx5_core_is_ecpf_esw_manager(esw->dev))
		return false;

	/* External host number starts with zero in device */
	if (controller == mlx5_esw_get_hpf_host_number(esw->dev) + 1)
		return true;

	esw_funcs = &esw->esw_funcs;
	for (i = 0; i < esw_funcs->num_spfs; i++) {
		if (controller == esw_funcs->spfs[i].host_number + 1)
			return true;
	}
	return false;
}

int esw_offloads_enable(struct mlx5_eswitch *esw)
{
	u8 mapping_id[MLX5_SW_IMAGE_GUID_MAX_BYTES];
	struct mlx5_devcom_match_attr attr = {};
	struct mapping_ctx *reg_c0_obj_pool;
	struct mlx5_vport *vport;
	unsigned long i;
	u8 id_len;
	int err;

	mutex_init(&esw->offloads.termtbl_mutex);
	mlx5_esw_adjacent_vhcas_setup(esw);

	err = mlx5_rdma_enable_roce(esw->dev);
	if (err)
		goto err_roce;

	/* SD devices defer metadata init until SD is ready and
	 * mlx5_sd_pf_num_get() can return the correct pf_num.
	 */
	if (!mlx5_get_sd(esw->dev)) {
		err = esw_offloads_metadata_init(esw);
		if (err)
			goto err_metadata;
	}

	err = esw_set_passing_vport_metadata(esw, true);
	if (err)
		goto err_vport_metadata;

	mlx5_query_nic_sw_system_image_guid(esw->dev, mapping_id, &id_len);

	reg_c0_obj_pool = mapping_create_for_id(mapping_id, id_len,
						MAPPING_TYPE_CHAIN,
						sizeof(struct mlx5_mapped_obj),
						ESW_REG_C0_USER_DATA_METADATA_MASK,
						true);

	if (IS_ERR(reg_c0_obj_pool)) {
		err = PTR_ERR(reg_c0_obj_pool);
		goto err_pool;
	}
	esw->offloads.reg_c0_obj_pool = reg_c0_obj_pool;

	err = esw_offloads_steering_init(esw);
	if (err)
		goto err_steering_init;

	if (esw->offloads_inactive)
		mlx5_esw_fdb_inactive(esw);
	else
		mlx5_esw_fdb_active(esw);

	/* Representor will control the vport link state */
	mlx5_esw_for_each_vf_vport(esw, i, vport, esw->esw_funcs.num_vfs)
		vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN;
	if (mlx5_core_ec_sriov_enabled(esw->dev))
		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs)
			vport->info.link_state = MLX5_VPORT_ADMIN_STATE_DOWN;

	/* Uplink vport rep must load first. */
	err = mlx5_esw_offloads_rep_load(esw, MLX5_VPORT_UPLINK);
	if (err)
		goto err_uplink;

	err = mlx5_eswitch_enable_pf_vf_vports(esw, MLX5_VPORT_UC_ADDR_CHANGE);
	if (err)
		goto err_vports;

	memcpy(attr.key.buf, mapping_id, id_len);
	attr.flags = MLX5_DEVCOM_MATCH_FLAGS_NS;
	attr.net = mlx5_core_net(esw->dev);
	mlx5_esw_offloads_devcom_init(esw, &attr);
	return 0;

err_vports:
	/* rollback to legacy, indicates don't unregister the uplink netdev */
	esw->dev->priv.flags |= MLX5_PRIV_FLAGS_SWITCH_LEGACY;
	mlx5_esw_offloads_rep_unload(esw, MLX5_VPORT_UPLINK);
err_uplink:
	esw_offloads_steering_cleanup(esw);
err_steering_init:
	mapping_destroy(reg_c0_obj_pool);
err_pool:
	esw_set_passing_vport_metadata(esw, false);
err_vport_metadata:
	esw_offloads_metadata_uninit(esw);
err_metadata:
	mlx5_rdma_disable_roce(esw->dev);
err_roce:
	mlx5_esw_adjacent_vhcas_cleanup(esw);
	mutex_destroy(&esw->offloads.termtbl_mutex);
	return err;
}

static int esw_offloads_stop(struct mlx5_eswitch *esw,
			     struct netlink_ext_ack *extack)
{
	int err;

	esw_mode_change(esw, MLX5_ESWITCH_LEGACY);

	/* If changing from switchdev to legacy mode without sriov enabled,
	 * no need to create legacy fdb.
	 */
	if (!mlx5_core_is_pf(esw->dev) || !mlx5_sriov_is_enabled(esw->dev))
		return 0;

	err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS);
	if (err)
		NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to legacy");

	return err;
}

void esw_offloads_disable(struct mlx5_eswitch *esw)
{
	mlx5_esw_offloads_devcom_cleanup(esw);
	mlx5_eswitch_disable_pf_vf_vports(esw);
	mlx5_esw_offloads_rep_unload(esw, MLX5_VPORT_UPLINK);
	esw_set_passing_vport_metadata(esw, false);
	esw_offloads_steering_cleanup(esw);
	mapping_destroy(esw->offloads.reg_c0_obj_pool);
	esw_offloads_metadata_uninit(esw);
	mlx5_rdma_disable_roce(esw->dev);
	mlx5_esw_adjacent_vhcas_cleanup(esw);
	/* must be done after vhcas cleanup to avoid adjacent vports connect */
	if (esw->offloads_inactive)
		mlx5_esw_fdb_active(esw); /* legacy mode always active */
	mutex_destroy(&esw->offloads.termtbl_mutex);
}

static int esw_mode_from_devlink(u16 mode, u16 *mlx5_mode)
{
	switch (mode) {
	case DEVLINK_ESWITCH_MODE_LEGACY:
		*mlx5_mode = MLX5_ESWITCH_LEGACY;
		break;
	case DEVLINK_ESWITCH_MODE_SWITCHDEV:
	case DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE:
		*mlx5_mode = MLX5_ESWITCH_OFFLOADS;
		break;
	default:
		return -EINVAL;
	}

	return 0;
}

static int esw_mode_to_devlink(struct mlx5_eswitch *esw, u16 *mode)
{
	switch (esw->mode) {
	case MLX5_ESWITCH_LEGACY:
		*mode = DEVLINK_ESWITCH_MODE_LEGACY;
		break;
	case MLX5_ESWITCH_OFFLOADS:
		if (esw->offloads_inactive)
			*mode = DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE;
		else
			*mode = DEVLINK_ESWITCH_MODE_SWITCHDEV;
		break;
	default:
		return -EINVAL;
	}

	return 0;
}

static int esw_inline_mode_from_devlink(u8 mode, u8 *mlx5_mode)
{
	switch (mode) {
	case DEVLINK_ESWITCH_INLINE_MODE_NONE:
		*mlx5_mode = MLX5_INLINE_MODE_NONE;
		break;
	case DEVLINK_ESWITCH_INLINE_MODE_LINK:
		*mlx5_mode = MLX5_INLINE_MODE_L2;
		break;
	case DEVLINK_ESWITCH_INLINE_MODE_NETWORK:
		*mlx5_mode = MLX5_INLINE_MODE_IP;
		break;
	case DEVLINK_ESWITCH_INLINE_MODE_TRANSPORT:
		*mlx5_mode = MLX5_INLINE_MODE_TCP_UDP;
		break;
	default:
		return -EINVAL;
	}

	return 0;
}

static int esw_inline_mode_to_devlink(u8 mlx5_mode, u8 *mode)
{
	switch (mlx5_mode) {
	case MLX5_INLINE_MODE_NONE:
		*mode = DEVLINK_ESWITCH_INLINE_MODE_NONE;
		break;
	case MLX5_INLINE_MODE_L2:
		*mode = DEVLINK_ESWITCH_INLINE_MODE_LINK;
		break;
	case MLX5_INLINE_MODE_IP:
		*mode = DEVLINK_ESWITCH_INLINE_MODE_NETWORK;
		break;
	case MLX5_INLINE_MODE_TCP_UDP:
		*mode = DEVLINK_ESWITCH_INLINE_MODE_TRANSPORT;
		break;
	default:
		return -EINVAL;
	}

	return 0;
}

int mlx5_eswitch_block_mode(struct mlx5_core_dev *dev)
{
	struct mlx5_eswitch *esw = dev->priv.eswitch;
	int err;

	if (!mlx5_esw_allowed(esw))
		return 0;

	/* Take TC into account */
	err = mlx5_esw_try_lock(esw);
	if (err < 0)
		return err;

	esw->offloads.num_block_mode++;
	mlx5_esw_unlock(esw);
	return 0;
}

void mlx5_eswitch_unblock_mode(struct mlx5_core_dev *dev)
{
	struct mlx5_eswitch *esw = dev->priv.eswitch;

	if (!mlx5_esw_allowed(esw))
		return;

	down_write(&esw->mode_lock);
	esw->offloads.num_block_mode--;
	up_write(&esw->mode_lock);
}

/* Returns false only when uplink netdev exists and its netns is different from
 * devlink's netns. True for all others so entering switchdev mode is allowed.
 */
static bool mlx5_devlink_netdev_netns_immutable_set(struct devlink *devlink,
						    bool immutable)
{
	struct mlx5_core_dev *mdev = devlink_priv(devlink);
	struct net_device *netdev;
	bool ret;

	netdev = mlx5_uplink_netdev_get(mdev);
	if (!netdev)
		return true;

	rtnl_lock();
	netdev->netns_immutable = immutable;
	ret = net_eq(dev_net(netdev), devlink_net(devlink));
	rtnl_unlock();

	mlx5_uplink_netdev_put(mdev, netdev);
	return ret;
}

/* Returns true when only changing between active and inactive switchdev mode */
static bool mlx5_devlink_switchdev_active_mode_change(struct mlx5_eswitch *esw,
						      u16 devlink_mode)
{
	/* current mode is not switchdev */
	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return false;

	/* new mode is not switchdev */
	if (devlink_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV &&
	    devlink_mode != DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE)
		return false;

	/* already inactive: no change in current state */
	if (devlink_mode == DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE &&
	    esw->offloads_inactive)
		return false;

	/* already active: no change in current state */
	if (devlink_mode == DEVLINK_ESWITCH_MODE_SWITCHDEV &&
	    !esw->offloads_inactive)
		return false;

	down_write(&esw->mode_lock);
	esw->offloads_inactive = !esw->offloads_inactive;
	esw->eswitch_operation_in_progress = true;
	up_write(&esw->mode_lock);

	if (esw->offloads_inactive)
		mlx5_esw_fdb_inactive(esw);
	else
		mlx5_esw_fdb_active(esw);

	down_write(&esw->mode_lock);
	esw->eswitch_operation_in_progress = false;
	up_write(&esw->mode_lock);
	return true;
}

#define MLX5_ESW_HOLD_TIMEOUT_MS 7000
#define MLX5_ESW_HOLD_RETRY_DELAY_MS 500

void mlx5_eswitch_safe_aux_devs_remove(struct mlx5_core_dev *dev)
{
	unsigned long timeout;
	bool hold_esw = true;

	/* Wait for any concurrent eswitch mode transition to complete. */
	if (!mlx5_esw_hold(dev)) {
		timeout = jiffies + msecs_to_jiffies(MLX5_ESW_HOLD_TIMEOUT_MS);
		while (!mlx5_esw_hold(dev)) {
			if (!time_before(jiffies, timeout)) {
				hold_esw = false;
				break;
			}
			msleep(MLX5_ESW_HOLD_RETRY_DELAY_MS);
		}
	}
	if (hold_esw) {
		if (mlx5_eswitch_mode(dev) == MLX5_ESWITCH_OFFLOADS)
			mlx5_core_reps_aux_devs_remove(dev);
		mlx5_esw_release(dev);
	}
}

int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode,
				  struct netlink_ext_ack *extack)
{
	u16 cur_mlx5_mode, mlx5_mode = 0;
	struct mlx5_eswitch *esw;
	int err = 0;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	if (mlx5_fw_reset_in_progress(esw->dev)) {
		NL_SET_ERR_MSG_MOD(extack, "Can't change eswitch mode during firmware reset");
		return -EBUSY;
	}

	if (esw_mode_from_devlink(mode, &mlx5_mode))
		return -EINVAL;

	/* Avoid try_lock, active/inactive mode change is not restricted */
	if (mlx5_devlink_switchdev_active_mode_change(esw, mode))
		return 0;

	mlx5_lag_disable_change(esw->dev);
	err = mlx5_esw_try_lock(esw);
	if (err < 0) {
		NL_SET_ERR_MSG_MOD(extack, "Can't change mode, E-Switch is busy");
		goto enable_lag;
	}
	cur_mlx5_mode = err;
	err = 0;

	if (cur_mlx5_mode == mlx5_mode)
		goto unlock;

	if (esw->offloads.num_block_mode) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Can't change eswitch mode when IPsec SA and/or policies are configured");
		err = -EOPNOTSUPP;
		goto unlock;
	}

	/* Keep mode_lock and reps_lock unnested. The operation flag excludes
	 * mode users while mode_lock is dropped before taking reps_lock.
	 */
	esw->eswitch_operation_in_progress = true;
	up_write(&esw->mode_lock);

	mlx5_esw_reps_block(esw);

	if (mlx5_mode == MLX5_ESWITCH_OFFLOADS &&
	    !mlx5_devlink_netdev_netns_immutable_set(devlink, true)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Can't change E-Switch mode to switchdev when netdev net namespace has diverged from the devlink's.");
		err = -EINVAL;
		goto skip;
	}

	if (mlx5_mode == MLX5_ESWITCH_LEGACY)
		esw->dev->priv.flags |= MLX5_PRIV_FLAGS_SWITCH_LEGACY;
	if (mlx5_mode == MLX5_ESWITCH_OFFLOADS)
		esw->dev->priv.flags &= ~MLX5_PRIV_FLAGS_SWITCH_LEGACY;
	mlx5_eswitch_disable_locked(esw);
	if (mlx5_mode == MLX5_ESWITCH_OFFLOADS) {
		if (mlx5_devlink_trap_get_num_active(esw->dev)) {
			NL_SET_ERR_MSG_MOD(extack,
					   "Can't change mode while devlink traps are active");
			err = -EOPNOTSUPP;
			goto skip;
		}
		esw->offloads_inactive =
			(mode == DEVLINK_ESWITCH_MODE_SWITCHDEV_INACTIVE);
		err = esw_offloads_start(esw, extack);
	} else if (mlx5_mode == MLX5_ESWITCH_LEGACY) {
		err = esw_offloads_stop(esw, extack);
	} else {
		err = -EINVAL;
	}

skip:
	if (mlx5_mode == MLX5_ESWITCH_OFFLOADS && err)
		mlx5_devlink_netdev_netns_immutable_set(devlink, false);
	/* Reconfiguration is done; drop reps_lock before taking mode_lock again
	 * to clear the operation flag.
	 */
	mlx5_esw_reps_unblock(esw);
	down_write(&esw->mode_lock);
	esw->eswitch_operation_in_progress = false;
unlock:
	mlx5_esw_unlock(esw);
enable_lag:
	mlx5_lag_enable_change(esw->dev);
	/* Shared FDB activation is creating LAG which is changing reps. */
	if (!err)
		mlx5_sd_eswitch_mode_set(esw->dev, mlx5_mode);
	return err;
}

int mlx5_devlink_eswitch_mode_get(struct devlink *devlink, u16 *mode)
{
	struct mlx5_eswitch *esw;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	return esw_mode_to_devlink(esw, mode);
}

static int mlx5_esw_vports_inline_set(struct mlx5_eswitch *esw, u8 mlx5_mode,
				      struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = esw->dev;
	struct mlx5_vport *vport;
	u16 err_vport_num = 0;
	unsigned long i;
	int err = 0;

	mlx5_esw_for_each_host_func_vport(esw, i, vport, esw->esw_funcs.num_vfs) {
		err = mlx5_modify_nic_vport_min_inline(dev, vport->vport, mlx5_mode);
		if (err) {
			err_vport_num = vport->vport;
			NL_SET_ERR_MSG_MOD(extack,
					   "Failed to set min inline on vport");
			goto revert_inline_mode;
		}
	}
	if (mlx5_core_ec_sriov_enabled(esw->dev)) {
		mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs) {
			err = mlx5_modify_nic_vport_min_inline(dev, vport->vport, mlx5_mode);
			if (err) {
				err_vport_num = vport->vport;
				NL_SET_ERR_MSG_MOD(extack,
						   "Failed to set min inline on vport");
				goto revert_ec_vf_inline_mode;
			}
		}
	}
	return 0;

revert_ec_vf_inline_mode:
	mlx5_esw_for_each_ec_vf_vport(esw, i, vport, esw->esw_funcs.num_ec_vfs) {
		if (vport->vport == err_vport_num)
			break;
		mlx5_modify_nic_vport_min_inline(dev,
						 vport->vport,
						 esw->offloads.inline_mode);
	}
revert_inline_mode:
	mlx5_esw_for_each_host_func_vport(esw, i, vport, esw->esw_funcs.num_vfs) {
		if (vport->vport == err_vport_num)
			break;
		mlx5_modify_nic_vport_min_inline(dev,
						 vport->vport,
						 esw->offloads.inline_mode);
	}
	return err;
}

int mlx5_devlink_eswitch_inline_mode_set(struct devlink *devlink, u8 mode,
					 struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = devlink_priv(devlink);
	struct mlx5_eswitch *esw;
	u8 mlx5_mode;
	int err;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	down_write(&esw->mode_lock);

	switch (MLX5_CAP_ETH(dev, wqe_inline_mode)) {
	case MLX5_CAP_INLINE_MODE_NOT_REQUIRED:
		if (mode == DEVLINK_ESWITCH_INLINE_MODE_NONE) {
			err = 0;
			goto out;
		}

		fallthrough;
	case MLX5_CAP_INLINE_MODE_L2:
		NL_SET_ERR_MSG_MOD(extack, "Inline mode can't be set");
		err = -EOPNOTSUPP;
		goto out;
	case MLX5_CAP_INLINE_MODE_VPORT_CONTEXT:
		break;
	}

	if (atomic64_read(&esw->offloads.num_flows) > 0) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Can't set inline mode when flows are configured");
		err = -EOPNOTSUPP;
		goto out;
	}

	err = esw_inline_mode_from_devlink(mode, &mlx5_mode);
	if (err)
		goto out;

	esw->eswitch_operation_in_progress = true;
	up_write(&esw->mode_lock);

	err = mlx5_esw_vports_inline_set(esw, mlx5_mode, extack);
	if (!err)
		esw->offloads.inline_mode = mlx5_mode;

	down_write(&esw->mode_lock);
	esw->eswitch_operation_in_progress = false;
	up_write(&esw->mode_lock);
	return 0;

out:
	up_write(&esw->mode_lock);
	return err;
}

int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode)
{
	struct mlx5_eswitch *esw;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	return esw_inline_mode_to_devlink(esw->offloads.inline_mode, mode);
}

bool mlx5_eswitch_block_encap(struct mlx5_core_dev *dev, bool from_fdb)
{
	struct mlx5_eswitch *esw = dev->priv.eswitch;
	enum devlink_eswitch_encap_mode encap;
	bool allow_tunnel = false;

	if (!mlx5_esw_allowed(esw))
		return true;

	down_write(&esw->mode_lock);
	encap = esw->offloads.encap;
	if (esw->mode == MLX5_ESWITCH_LEGACY ||
	    (encap == DEVLINK_ESWITCH_ENCAP_MODE_NONE && !from_fdb)) {
		allow_tunnel = true;
		esw->offloads.num_block_encap++;
	}
	up_write(&esw->mode_lock);

	return allow_tunnel;
}

void mlx5_eswitch_unblock_encap(struct mlx5_core_dev *dev)
{
	struct mlx5_eswitch *esw = dev->priv.eswitch;

	if (!mlx5_esw_allowed(esw))
		return;

	down_write(&esw->mode_lock);
	esw->offloads.num_block_encap--;
	up_write(&esw->mode_lock);
}

int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink,
					enum devlink_eswitch_encap_mode encap,
					struct netlink_ext_ack *extack)
{
	struct mlx5_core_dev *dev = devlink_priv(devlink);
	struct mlx5_eswitch *esw;
	int err = 0;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	down_write(&esw->mode_lock);

	if (encap != DEVLINK_ESWITCH_ENCAP_MODE_NONE &&
	    (!MLX5_CAP_ESW_FLOWTABLE_FDB(dev, reformat) ||
	     !MLX5_CAP_ESW_FLOWTABLE_FDB(dev, decap))) {
		err = -EOPNOTSUPP;
		goto unlock;
	}

	if (encap && encap != DEVLINK_ESWITCH_ENCAP_MODE_BASIC) {
		err = -EOPNOTSUPP;
		goto unlock;
	}

	if (esw->mode == MLX5_ESWITCH_LEGACY) {
		esw->offloads.encap = encap;
		goto unlock;
	}

	if (esw->offloads.encap == encap)
		goto unlock;

	if (atomic64_read(&esw->offloads.num_flows) > 0) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Can't set encapsulation when flows are configured");
		err = -EOPNOTSUPP;
		goto unlock;
	}

	if (esw->offloads.num_block_encap) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Can't set encapsulation when IPsec SA and/or policies are configured");
		err = -EOPNOTSUPP;
		goto unlock;
	}

	esw->eswitch_operation_in_progress = true;
	up_write(&esw->mode_lock);

	esw_destroy_offloads_fdb_tables(esw);

	esw->offloads.encap = encap;

	err = esw_create_offloads_fdb_tables(esw);

	if (err) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Failed re-creating fast FDB table");
		esw->offloads.encap = !encap;
		(void)esw_create_offloads_fdb_tables(esw);
	}

	down_write(&esw->mode_lock);
	esw->eswitch_operation_in_progress = false;

unlock:
	up_write(&esw->mode_lock);
	return err;
}

int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink,
					enum devlink_eswitch_encap_mode *encap)
{
	struct mlx5_eswitch *esw;

	esw = mlx5_devlink_eswitch_get(devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	*encap = esw->offloads.encap;
	return 0;
}

static bool
mlx5_eswitch_vport_has_rep(const struct mlx5_eswitch *esw, u16 vport_num)
{
	/* Currently, only ECPF based device has representor for host PF. */
	if (vport_num == MLX5_VPORT_HOST_PF &&
	    (!mlx5_core_is_ecpf_esw_manager(esw->dev) ||
	     !mlx5_esw_host_functions_enabled(esw->dev)))
		return false;

	if (vport_num == MLX5_VPORT_ECPF &&
	    !mlx5_ecpf_vport_exists(esw->dev))
		return false;

	return true;
}

static void
mlx5_eswitch_register_vport_reps_blocked(struct mlx5_eswitch *esw,
					 const struct mlx5_eswitch_rep_ops *ops,
					 u8 rep_type)
{
	struct mlx5_eswitch_rep_data *rep_data;
	struct mlx5_eswitch_rep *rep;
	unsigned long i;

	esw->offloads.rep_ops[rep_type] = ops;
	mlx5_esw_for_each_rep(esw, i, rep) {
		if (likely(mlx5_eswitch_vport_has_rep(esw, rep->vport))) {
			rep->esw = esw;
			rep_data = &rep->rep_data[rep_type];
			atomic_set(&rep_data->state, REP_REGISTERED);
		}
	}
}

static void mlx5_eswitch_reload_reps_blocked(struct mlx5_eswitch *esw)
{
	struct mlx5_eswitch_rep *uplink;
	struct mlx5_vport *vport;
	bool newly_loaded;
	unsigned long i;

	if (esw->mode != MLX5_ESWITCH_OFFLOADS)
		return;

	uplink = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
	if (__esw_offloads_load_rep(esw, uplink, REP_ETH, &newly_loaded))
		return;
	if (mlx5_sd_is_primary(esw->dev) &&
	    __esw_offloads_load_rep(esw, uplink, REP_IB, NULL)) {
		if (newly_loaded)
			__esw_offloads_unload_rep(esw, uplink, REP_ETH);
		return;
	}

	if (mlx5_get_sd(esw->dev) && !mlx5_lag_is_active(esw->dev))
		return;

	mlx5_esw_for_each_vport(esw, i, vport) {
		if (!vport)
			continue;
		if (!vport->enabled)
			continue;
		if (vport->vport == MLX5_VPORT_UPLINK)
			continue;
		if (!mlx5_eswitch_vport_has_rep(esw, vport->vport))
			continue;

		mlx5_esw_offloads_rep_load(esw, vport->vport);
	}
}

static void mlx5_eswitch_reload_reps(struct mlx5_eswitch *esw)
{
	mlx5_esw_reps_block(esw);
	mlx5_eswitch_reload_reps_blocked(esw);
	mlx5_esw_reps_unblock(esw);
}

static void
mlx5_eswitch_register_vport_reps_locked(struct mlx5_eswitch *esw,
					const struct mlx5_eswitch_rep_ops *ops,
					u8 rep_type, bool nested)
{
	if (nested)
		mlx5_esw_reps_block_nested(esw);
	else
		mlx5_esw_reps_block(esw);
	mlx5_eswitch_register_vport_reps_blocked(esw, ops, rep_type);
	mlx5_esw_reps_unblock(esw);

	mlx5_esw_add_work(esw, mlx5_eswitch_reload_reps, GFP_KERNEL);
}

void mlx5_eswitch_register_vport_reps(struct mlx5_eswitch *esw,
				      const struct mlx5_eswitch_rep_ops *ops,
				      u8 rep_type)
{
	mlx5_eswitch_register_vport_reps_locked(esw, ops, rep_type, false);
}
EXPORT_SYMBOL(mlx5_eswitch_register_vport_reps);

void
mlx5_eswitch_register_vport_reps_nested(struct mlx5_eswitch *esw,
					const struct mlx5_eswitch_rep_ops *ops,
					u8 rep_type)
{
	mlx5_eswitch_register_vport_reps_locked(esw, ops, rep_type, true);
}
EXPORT_SYMBOL(mlx5_eswitch_register_vport_reps_nested);

static void
mlx5_eswitch_unregister_vport_reps_blocked(struct mlx5_eswitch *esw,
					   u8 rep_type)
{
	struct mlx5_eswitch_rep *rep;
	unsigned long i;

	__unload_reps_all_vport(esw, rep_type);

	mlx5_esw_for_each_rep(esw, i, rep)
		atomic_set(&rep->rep_data[rep_type].state, REP_UNREGISTERED);

	esw->offloads.rep_ops[rep_type] = NULL;
}

static void
mlx5_eswitch_unregister_vport_reps_locked(struct mlx5_eswitch *esw,
					  u8 rep_type, bool nested)
{
	if (nested)
		mlx5_esw_reps_block_nested(esw);
	else
		mlx5_esw_reps_block(esw);
	mlx5_eswitch_unregister_vport_reps_blocked(esw, rep_type);
	mlx5_esw_reps_unblock(esw);
}

void mlx5_eswitch_unregister_vport_reps(struct mlx5_eswitch *esw, u8 rep_type)
{
	mlx5_eswitch_unregister_vport_reps_locked(esw, rep_type, false);
}
EXPORT_SYMBOL(mlx5_eswitch_unregister_vport_reps);

void mlx5_eswitch_unregister_vport_reps_nested(struct mlx5_eswitch *esw,
					       u8 rep_type)
{
	mlx5_eswitch_unregister_vport_reps_locked(esw, rep_type, true);
}
EXPORT_SYMBOL(mlx5_eswitch_unregister_vport_reps_nested);

void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type)
{
	struct mlx5_core_dev *primary = mlx5_sd_get_primary(esw->dev);
	struct mlx5_eswitch_rep *rep;

	if (primary)
		esw = primary->priv.eswitch;
	rep = mlx5_eswitch_get_rep(esw, MLX5_VPORT_UPLINK);
	return rep->rep_data[rep_type].priv;
}

void *mlx5_eswitch_get_proto_dev(struct mlx5_eswitch *esw,
				 u16 vport,
				 u8 rep_type)
{
	struct mlx5_eswitch_rep *rep;

	rep = mlx5_eswitch_get_rep(esw, vport);

	if (atomic_read(&rep->rep_data[rep_type].state) == REP_LOADED &&
	    esw->offloads.rep_ops[rep_type]->get_proto_dev)
		return esw->offloads.rep_ops[rep_type]->get_proto_dev(rep);
	return NULL;
}
EXPORT_SYMBOL(mlx5_eswitch_get_proto_dev);

void *mlx5_eswitch_uplink_get_proto_dev(struct mlx5_eswitch *esw, u8 rep_type)
{
	struct mlx5_core_dev *primary = mlx5_sd_get_primary(esw->dev);

	if (primary)
		esw = primary->priv.eswitch;

	return mlx5_eswitch_get_proto_dev(esw, MLX5_VPORT_UPLINK, rep_type);
}
EXPORT_SYMBOL(mlx5_eswitch_uplink_get_proto_dev);

struct mlx5_eswitch_rep *mlx5_eswitch_vport_rep(struct mlx5_eswitch *esw,
						u16 vport)
{
	return mlx5_eswitch_get_rep(esw, vport);
}
EXPORT_SYMBOL(mlx5_eswitch_vport_rep);

bool mlx5_eswitch_reg_c1_loopback_enabled(const struct mlx5_eswitch *esw)
{
	return !!(esw->flags & MLX5_ESWITCH_REG_C1_LOOPBACK_ENABLED);
}
EXPORT_SYMBOL(mlx5_eswitch_reg_c1_loopback_enabled);

bool mlx5_eswitch_vport_match_metadata_enabled(const struct mlx5_eswitch *esw)
{
	return !!(esw->flags & MLX5_ESWITCH_VPORT_MATCH_METADATA);
}
EXPORT_SYMBOL(mlx5_eswitch_vport_match_metadata_enabled);

u32 mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw,
					      u16 vport_num)
{
	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);

	if (WARN_ON_ONCE(IS_ERR(vport)))
		return 0;

	return vport->metadata << (32 - ESW_SOURCE_PORT_METADATA_BITS);
}
EXPORT_SYMBOL(mlx5_eswitch_get_vport_metadata_for_match);

int mlx5_esw_vport_vhca_id_map(struct mlx5_eswitch *esw,
			       struct mlx5_vport *vport)
{
	u16 *old_entry, *vhca_map_entry, vhca_id;

	if (WARN_ONCE(MLX5_VPORT_INVAL_VHCA_ID(vport),
		      "vport %d vhca_id is not set", vport->vport)) {
		int err;

		err = mlx5_vport_get_vhca_id(vport->dev, vport->vport,
					     &vhca_id);
		if (err)
			return err;
		vport->vhca_id = vhca_id;
	}

	vhca_id = vport->vhca_id;
	vhca_map_entry = kmalloc_obj(*vhca_map_entry);
	if (!vhca_map_entry)
		return -ENOMEM;

	*vhca_map_entry = vport->vport;
	old_entry = xa_store(&esw->offloads.vhca_map, vhca_id, vhca_map_entry, GFP_KERNEL);
	if (xa_is_err(old_entry)) {
		kfree(vhca_map_entry);
		return xa_err(old_entry);
	}
	kfree(old_entry);
	return 0;
}

void mlx5_esw_vport_vhca_id_unmap(struct mlx5_eswitch *esw,
				  struct mlx5_vport *vport)
{
	u16 *vhca_map_entry;

	vhca_map_entry = xa_erase(&esw->offloads.vhca_map, vport->vhca_id);
	kfree(vhca_map_entry);
}

int mlx5_eswitch_vhca_id_to_vport(struct mlx5_eswitch *esw, u16 vhca_id, u16 *vport_num)
{
	u16 *res = xa_load(&esw->offloads.vhca_map, vhca_id);

	if (!res)
		return -ENOENT;

	*vport_num = *res;
	return 0;
}

u32 mlx5_eswitch_get_vport_metadata_for_set(struct mlx5_eswitch *esw,
					    u16 vport_num)
{
	struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num);

	if (WARN_ON_ONCE(IS_ERR(vport)))
		return 0;

	return vport->metadata;
}
EXPORT_SYMBOL(mlx5_eswitch_get_vport_metadata_for_set);

int mlx5_devlink_port_fn_hw_addr_get(struct devlink_port *port,
				     u8 *hw_addr, int *hw_addr_len,
				     struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);

	mutex_lock(&esw->state_lock);

	mlx5_query_nic_vport_mac_address(esw->dev, vport->vport, true,
					 vport->info.mac);
	ether_addr_copy(hw_addr, vport->info.mac);
	*hw_addr_len = ETH_ALEN;
	mutex_unlock(&esw->state_lock);
	return 0;
}

int mlx5_devlink_port_fn_hw_addr_set(struct devlink_port *port,
				     const u8 *hw_addr, int hw_addr_len,
				     struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);

	return mlx5_eswitch_set_vport_mac(esw, vport->vport, hw_addr);
}

int mlx5_devlink_port_fn_migratable_get(struct devlink_port *port, bool *is_enabled,
					struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);

	if (!MLX5_CAP_GEN(esw->dev, migration)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support migration");
		return -EOPNOTSUPP;
	}

	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support vport group management");
		return -EOPNOTSUPP;
	}

	mutex_lock(&esw->state_lock);
	*is_enabled = vport->info.mig_enabled;
	mutex_unlock(&esw->state_lock);
	return 0;
}

int mlx5_devlink_port_fn_migratable_set(struct devlink_port *port, bool enable,
					struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out);
	void *query_ctx;
	void *hca_caps;
	int err;

	if (!MLX5_CAP_GEN(esw->dev, migration)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support migration");
		return -EOPNOTSUPP;
	}

	if (!MLX5_CAP_GEN(esw->dev, vhca_resource_manager)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support VHCA management");
		return -EOPNOTSUPP;
	}

	mutex_lock(&esw->state_lock);

	if (vport->info.mig_enabled == enable) {
		err = 0;
		goto out;
	}

	query_ctx = kzalloc(query_out_sz, GFP_KERNEL);
	if (!query_ctx) {
		err = -ENOMEM;
		goto out;
	}

	err = mlx5_vport_get_other_func_cap(esw->dev, vport->vport, query_ctx,
					    MLX5_CAP_GENERAL_2);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps");
		goto out_free;
	}

	hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability);
	MLX5_SET(cmd_hca_cap_2, hca_caps, migratable, enable);

	err = mlx5_vport_set_other_func_cap(esw->dev, hca_caps, vport->vport,
					    MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA migratable cap");
		goto out_free;
	}

	vport->info.mig_enabled = enable;

out_free:
	kfree(query_ctx);
out:
	mutex_unlock(&esw->state_lock);
	return err;
}

int mlx5_devlink_port_fn_roce_get(struct devlink_port *port, bool *is_enabled,
				  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);

	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support vport group management");
		return -EOPNOTSUPP;
	}

	mutex_lock(&esw->state_lock);
	*is_enabled = vport->info.roce_enabled;
	mutex_unlock(&esw->state_lock);
	return 0;
}

int mlx5_devlink_port_fn_roce_set(struct devlink_port *port, bool enable,
				  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out);
	u16 vport_num = vport->vport;
	void *query_ctx;
	void *hca_caps;
	int err;

	if (!MLX5_CAP_GEN(esw->dev, vhca_resource_manager)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support VHCA management");
		return -EOPNOTSUPP;
	}

	mutex_lock(&esw->state_lock);

	if (vport->info.roce_enabled == enable) {
		err = 0;
		goto out;
	}

	query_ctx = kzalloc(query_out_sz, GFP_KERNEL);
	if (!query_ctx) {
		err = -ENOMEM;
		goto out;
	}

	err = mlx5_vport_get_other_func_cap(esw->dev, vport_num, query_ctx,
					    MLX5_CAP_GENERAL);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps");
		goto out_free;
	}

	hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability);
	MLX5_SET(cmd_hca_cap, hca_caps, roce, enable);

	err = mlx5_vport_set_other_func_general_cap(esw->dev, hca_caps,
						    vport_num);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA roce cap");
		goto out_free;
	}

	vport->info.roce_enabled = enable;

out_free:
	kfree(query_ctx);
out:
	mutex_unlock(&esw->state_lock);
	return err;
}

int mlx5_devlink_pf_port_fn_state_get(struct devlink_port *port,
				      enum devlink_port_fn_state *state,
				      enum devlink_port_fn_opstate *opstate,
				      struct netlink_ext_ack *extack)
{
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	struct mlx5_eswitch *esw = vport->dev->priv.eswitch;
	const u32 *query_out;
	bool pf_disabled;

	if (mlx5_eswitch_is_vf_vport(esw, vport->vport)) {
		NL_SET_ERR_MSG_MOD(extack, "State get is not supported for VF");
		return -EOPNOTSUPP;
	}

	*state = vport->pf_activated ?
		 DEVLINK_PORT_FN_STATE_ACTIVE : DEVLINK_PORT_FN_STATE_INACTIVE;

	query_out = mlx5_esw_query_functions(vport->dev);
	if (IS_ERR(query_out))
		return PTR_ERR(query_out);

	if (vport->vport == MLX5_VPORT_HOST_PF) {
		struct mlx5_esw_pf_info host_pf_info;

		host_pf_info = mlx5_esw_get_host_pf_info(vport->dev,
							 query_out);
		pf_disabled = host_pf_info.pf_disabled;
	} else {
		pf_disabled = mlx5_esw_get_spf_disabled(vport->dev, query_out,
							vport->vhca_id);
	}

	*opstate = pf_disabled ? DEVLINK_PORT_FN_OPSTATE_DETACHED :
				 DEVLINK_PORT_FN_OPSTATE_ATTACHED;

	kvfree(query_out);
	return 0;
}

int mlx5_devlink_pf_port_fn_state_set(struct devlink_port *port,
				      enum devlink_port_fn_state state,
				      struct netlink_ext_ack *extack)
{
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	struct mlx5_eswitch *esw = vport->dev->priv.eswitch;
	struct mlx5_core_dev *dev;

	if (mlx5_eswitch_is_vf_vport(esw, vport->vport)) {
		NL_SET_ERR_MSG_MOD(extack, "State set is not supported for VF");
		return -EOPNOTSUPP;
	}

	dev = vport->dev;

	switch (state) {
	case DEVLINK_PORT_FN_STATE_ACTIVE:
		return mlx5_esw_pf_enable_hca(dev, vport->vport);
	case DEVLINK_PORT_FN_STATE_INACTIVE:
		return mlx5_esw_pf_disable_hca(dev, vport->vport);
	default:
		return -EOPNOTSUPP;
	}
}

int
mlx5_eswitch_restore_ipsec_rule(struct mlx5_eswitch *esw, struct mlx5_flow_handle *rule,
				struct mlx5_esw_flow_attr *esw_attr, int attr_idx)
{
	struct mlx5_flow_destination new_dest = {};
	struct mlx5_flow_destination old_dest = {};

	if (!esw_setup_uplink_fwd_ipsec_needed(esw, esw_attr, attr_idx))
		return 0;

	esw_setup_dest_fwd_ipsec(&old_dest, NULL, esw, esw_attr, attr_idx, 0, false);
	esw_setup_dest_fwd_vport(&new_dest, NULL, esw, esw_attr, attr_idx, 0, false);

	return mlx5_modify_rule_destination(rule, &new_dest, &old_dest);
}

#ifdef CONFIG_XFRM_OFFLOAD
int mlx5_devlink_port_fn_ipsec_crypto_get(struct devlink_port *port, bool *is_enabled,
					  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw;
	struct mlx5_vport *vport;
	int err = 0;

	esw = mlx5_devlink_eswitch_get(port->devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	if (!mlx5_esw_ipsec_vf_offload_supported(esw->dev)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support IPSec crypto");
		return -EOPNOTSUPP;
	}

	vport = mlx5_devlink_port_vport_get(port);

	mutex_lock(&esw->state_lock);
	if (!vport->enabled) {
		err = -EOPNOTSUPP;
		goto unlock;
	}

	*is_enabled = vport->info.ipsec_crypto_enabled;
unlock:
	mutex_unlock(&esw->state_lock);
	return err;
}

int mlx5_devlink_port_fn_ipsec_crypto_set(struct devlink_port *port, bool enable,
					  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw;
	struct mlx5_vport *vport;
	u16 vport_num;
	int err;

	esw = mlx5_devlink_eswitch_get(port->devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	vport_num = mlx5_esw_devlink_port_index_to_vport_num(port->index);
	err = mlx5_esw_ipsec_vf_crypto_offload_supported(esw->dev, vport_num);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support IPsec crypto");
		return err;
	}

	vport = mlx5_devlink_port_vport_get(port);

	mutex_lock(&esw->state_lock);
	if (!vport->enabled) {
		err = -EOPNOTSUPP;
		NL_SET_ERR_MSG_MOD(extack, "Eswitch vport is disabled");
		goto unlock;
	}

	if (vport->info.ipsec_crypto_enabled == enable)
		goto unlock;

	if (!esw->enabled_ipsec_vf_count && esw->dev->num_ipsec_offloads) {
		err = -EBUSY;
		goto unlock;
	}

	err = mlx5_esw_ipsec_vf_crypto_offload_set(esw, vport, enable);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed to set IPsec crypto");
		goto unlock;
	}

	vport->info.ipsec_crypto_enabled = enable;
	if (enable)
		esw->enabled_ipsec_vf_count++;
	else
		esw->enabled_ipsec_vf_count--;
unlock:
	mutex_unlock(&esw->state_lock);
	return err;
}

int mlx5_devlink_port_fn_ipsec_packet_get(struct devlink_port *port, bool *is_enabled,
					  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw;
	struct mlx5_vport *vport;
	int err = 0;

	esw = mlx5_devlink_eswitch_get(port->devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	if (!mlx5_esw_ipsec_vf_offload_supported(esw->dev)) {
		NL_SET_ERR_MSG_MOD(extack, "Device doesn't support IPsec packet");
		return -EOPNOTSUPP;
	}

	vport = mlx5_devlink_port_vport_get(port);

	mutex_lock(&esw->state_lock);
	if (!vport->enabled) {
		err = -EOPNOTSUPP;
		goto unlock;
	}

	*is_enabled = vport->info.ipsec_packet_enabled;
unlock:
	mutex_unlock(&esw->state_lock);
	return err;
}

int mlx5_devlink_port_fn_ipsec_packet_set(struct devlink_port *port,
					  bool enable,
					  struct netlink_ext_ack *extack)
{
	struct mlx5_eswitch *esw;
	struct mlx5_vport *vport;
	u16 vport_num;
	int err;

	esw = mlx5_devlink_eswitch_get(port->devlink);
	if (IS_ERR(esw))
		return PTR_ERR(esw);

	vport_num = mlx5_esw_devlink_port_index_to_vport_num(port->index);
	err = mlx5_esw_ipsec_vf_packet_offload_supported(esw->dev, vport_num);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support IPsec packet mode");
		return err;
	}

	vport = mlx5_devlink_port_vport_get(port);
	mutex_lock(&esw->state_lock);
	if (!vport->enabled) {
		err = -EOPNOTSUPP;
		NL_SET_ERR_MSG_MOD(extack, "Eswitch vport is disabled");
		goto unlock;
	}

	if (vport->info.ipsec_packet_enabled == enable)
		goto unlock;

	if (!esw->enabled_ipsec_vf_count && esw->dev->num_ipsec_offloads) {
		err = -EBUSY;
		goto unlock;
	}

	err = mlx5_esw_ipsec_vf_packet_offload_set(esw, vport, enable);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Failed to set IPsec packet mode");
		goto unlock;
	}

	vport->info.ipsec_packet_enabled = enable;
	if (enable)
		esw->enabled_ipsec_vf_count++;
	else
		esw->enabled_ipsec_vf_count--;
unlock:
	mutex_unlock(&esw->state_lock);
	return err;
}
#endif /* CONFIG_XFRM_OFFLOAD */

int
mlx5_devlink_port_fn_max_io_eqs_get(struct devlink_port *port, u32 *max_io_eqs,
				    struct netlink_ext_ack *extack)
{
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out);
	u16 vport_num = vport->vport;
	struct mlx5_eswitch *esw;
	void *query_ctx;
	void *hca_caps;
	u32 max_eqs;
	int err;

	esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support vport group management");
		return -EOPNOTSUPP;
	}

	if (!MLX5_CAP_GEN_2(esw->dev, max_num_eqs_24b)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support getting the max number of EQs");
		return -EOPNOTSUPP;
	}

	query_ctx = kzalloc(query_out_sz, GFP_KERNEL);
	if (!query_ctx)
		return -ENOMEM;

	mutex_lock(&esw->state_lock);
	err = mlx5_vport_get_other_func_cap(esw->dev, vport_num, query_ctx,
					    MLX5_CAP_GENERAL_2);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps");
		goto out;
	}

	hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability);
	max_eqs = MLX5_GET(cmd_hca_cap_2, hca_caps, max_num_eqs_24b);
	if (max_eqs < MLX5_ESW_MAX_CTRL_EQS)
		*max_io_eqs = 0;
	else
		*max_io_eqs = max_eqs - MLX5_ESW_MAX_CTRL_EQS;
out:
	mutex_unlock(&esw->state_lock);
	kfree(query_ctx);
	return err;
}

int
mlx5_devlink_port_fn_max_io_eqs_set(struct devlink_port *port, u32 max_io_eqs,
				    struct netlink_ext_ack *extack)
{
	struct mlx5_vport *vport = mlx5_devlink_port_vport_get(port);
	int query_out_sz = MLX5_ST_SZ_BYTES(query_hca_cap_out);
	u16 vport_num = vport->vport;
	struct mlx5_eswitch *esw;
	void *query_ctx;
	void *hca_caps;
	u16 max_eqs;
	int err;

	esw = mlx5_devlink_eswitch_nocheck_get(port->devlink);
	if (!MLX5_CAP_GEN(esw->dev, vhca_resource_manager)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support VHCA management");
		return -EOPNOTSUPP;
	}

	if (!MLX5_CAP_GEN_2(esw->dev, max_num_eqs_24b)) {
		NL_SET_ERR_MSG_MOD(extack,
				   "Device doesn't support changing the max number of EQs");
		return -EOPNOTSUPP;
	}

	if (check_add_overflow(max_io_eqs, MLX5_ESW_MAX_CTRL_EQS, &max_eqs)) {
		NL_SET_ERR_MSG_MOD(extack, "Supplied value out of range");
		return -EINVAL;
	}

	query_ctx = kzalloc(query_out_sz, GFP_KERNEL);
	if (!query_ctx)
		return -ENOMEM;

	mutex_lock(&esw->state_lock);
	err = mlx5_vport_get_other_func_cap(esw->dev, vport_num, query_ctx,
					    MLX5_CAP_GENERAL_2);
	if (err) {
		NL_SET_ERR_MSG_MOD(extack, "Failed getting HCA caps");
		goto out;
	}

	hca_caps = MLX5_ADDR_OF(query_hca_cap_out, query_ctx, capability);
	MLX5_SET(cmd_hca_cap_2, hca_caps, max_num_eqs_24b, max_eqs);

	if (mlx5_esw_is_sf_vport(esw, vport_num))
		MLX5_SET(cmd_hca_cap_2, hca_caps, sf_eq_usage, 1);

	err = mlx5_vport_set_other_func_cap(esw->dev, hca_caps, vport_num,
					    MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE2);
	if (err)
		NL_SET_ERR_MSG_MOD(extack, "Failed setting HCA caps");
	vport->max_eqs_set = true;
out:
	mutex_unlock(&esw->state_lock);
	kfree(query_ctx);
	return err;
}

int
mlx5_devlink_port_fn_max_io_eqs_set_sf_default(struct devlink_port *port,
					       struct netlink_ext_ack *extack)
{
	return mlx5_devlink_port_fn_max_io_eqs_set(port,
						   MLX5_ESW_DEFAULT_SF_COMP_EQS,
						   extack);
}