Merge tag 'mlx5-fixes-2023-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5 fixes 2023-03-15

This series provides bug fixes to mlx5 driver.

* tag 'mlx5-fixes-2023-03-15' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5e: TC, Remove error message log print
  net/mlx5e: TC, fix cloned flow attribute
  net/mlx5e: TC, fix missing error code
  net/sched: TC, fix raw counter initialization
  net/mlx5e: Lower maximum allowed MTU in XSK to match XDP prerequisites
  net/mlx5: Set BREAK_FW_WAIT flag first when removing driver
  net/mlx5e: kTLS, Fix missing error unwind on unsupported cipher type
  net/mlx5e: Fix cleanup null-ptr deref on encap lock
  net/mlx5: E-switch, Fix missing set of split_count when forward to ovs internal port
  net/mlx5: E-switch, Fix wrong usage of source port rewrite in split rules
  net/mlx5: Disable eswitch before waiting for VF pages
  net/mlx5: Fix setting ec_function bit in MANAGE_PAGES
  net/mlx5e: Don't cache tunnel offloads capability
  net/mlx5e: Fix macsec ASO context alignment
====================

Link: https://lore.kernel.org/r/20230315225847.360083-1-saeed@kernel.orgSigned-off-by: Jakub Kicinski <kuba@kernel.org>

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

@@ -313,7 +313,6 @@ struct mlx5e_params {
 		} channel;
 	} mqprio;
 	bool rx_cqe_compress_def;
-	bool tunneled_offload_en;
 	struct dim_cq_moder rx_cq_moderation;
 	struct dim_cq_moder tx_cq_moderation;
 	struct mlx5e_packet_merge_param packet_merge;

drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/police.c

@@ -178,7 +178,6 @@ tc_act_police_stats(struct mlx5e_priv *priv,
 	meter = mlx5e_tc_meter_get(priv->mdev, &params);
 	if (IS_ERR(meter)) {
 		NL_SET_ERR_MSG_MOD(fl_act->extack, "Failed to get flow meter");
-		mlx5_core_err(priv->mdev, "Failed to get flow meter %d\n", params.index);
 		return PTR_ERR(meter);
 	}
 

drivers/net/ethernet/mellanox/mlx5/core/en/tc/act_stats.c

@@ -64,6 +64,7 @@ mlx5e_tc_act_stats_add(struct mlx5e_tc_act_stats_handle *handle,
 {
 	struct mlx5e_tc_act_stats *act_stats, *old_act_stats;
 	struct rhashtable *ht = &handle->ht;
+	u64 lastused;
 	int err = 0;
 
 	act_stats = kvzalloc(sizeof(*act_stats), GFP_KERNEL);
@@ -73,6 +74,10 @@ mlx5e_tc_act_stats_add(struct mlx5e_tc_act_stats_handle *handle,
 	act_stats->tc_act_cookie = act_cookie;
 	act_stats->counter = counter;
 
+	mlx5_fc_query_cached_raw(counter,
+				 &act_stats->lastbytes,
+				 &act_stats->lastpackets, &lastused);
+
 	rcu_read_lock();
 	old_act_stats = rhashtable_lookup_get_insert_fast(ht,
 							  &act_stats->hash,

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c

@@ -621,15 +621,6 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk,
 	if (unlikely(!priv_rx))
 		return -ENOMEM;
 
-	dek = mlx5_ktls_create_key(priv->tls->dek_pool, crypto_info);
-	if (IS_ERR(dek)) {
-		err = PTR_ERR(dek);
-		goto err_create_key;
-	}
-	priv_rx->dek = dek;
-
-	INIT_LIST_HEAD(&priv_rx->list);
-	spin_lock_init(&priv_rx->lock);
 	switch (crypto_info->cipher_type) {
 	case TLS_CIPHER_AES_GCM_128:
 		priv_rx->crypto_info.crypto_info_128 =
@@ -642,9 +633,20 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk,
 	default:
 		WARN_ONCE(1, "Unsupported cipher type %u\n",
 			  crypto_info->cipher_type);
-		return -EOPNOTSUPP;
+		err = -EOPNOTSUPP;
+		goto err_cipher_type;
 	}
 
+	dek = mlx5_ktls_create_key(priv->tls->dek_pool, crypto_info);
+	if (IS_ERR(dek)) {
+		err = PTR_ERR(dek);
+		goto err_cipher_type;
+	}
+	priv_rx->dek = dek;
+
+	INIT_LIST_HEAD(&priv_rx->list);
+	spin_lock_init(&priv_rx->lock);
+
 	rxq = mlx5e_ktls_sk_get_rxq(sk);
 	priv_rx->rxq = rxq;
 	priv_rx->sk = sk;
@@ -677,7 +679,7 @@ int mlx5e_ktls_add_rx(struct net_device *netdev, struct sock *sk,
 	mlx5e_tir_destroy(&priv_rx->tir);
 err_create_tir:
 	mlx5_ktls_destroy_key(priv->tls->dek_pool, priv_rx->dek);
-err_create_key:
+err_cipher_type:
 	kfree(priv_rx);
 	return err;
 }

drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c

@@ -469,14 +469,6 @@ int mlx5e_ktls_add_tx(struct net_device *netdev, struct sock *sk,
 	if (IS_ERR(priv_tx))
 		return PTR_ERR(priv_tx);
 
-	dek = mlx5_ktls_create_key(priv->tls->dek_pool, crypto_info);
-	if (IS_ERR(dek)) {
-		err = PTR_ERR(dek);
-		goto err_create_key;
-	}
-	priv_tx->dek = dek;
-
-	priv_tx->expected_seq = start_offload_tcp_sn;
 	switch (crypto_info->cipher_type) {
 	case TLS_CIPHER_AES_GCM_128:
 		priv_tx->crypto_info.crypto_info_128 =
@@ -489,8 +481,18 @@ int mlx5e_ktls_add_tx(struct net_device *netdev, struct sock *sk,
 	default:
 		WARN_ONCE(1, "Unsupported cipher type %u\n",
 			  crypto_info->cipher_type);
-		return -EOPNOTSUPP;
+		err = -EOPNOTSUPP;
+		goto err_pool_push;
 	}
+
+	dek = mlx5_ktls_create_key(priv->tls->dek_pool, crypto_info);
+	if (IS_ERR(dek)) {
+		err = PTR_ERR(dek);
+		goto err_pool_push;
+	}
+
+	priv_tx->dek = dek;
+	priv_tx->expected_seq = start_offload_tcp_sn;
 	priv_tx->tx_ctx = tls_offload_ctx_tx(tls_ctx);
 
 	mlx5e_set_ktls_tx_priv_ctx(tls_ctx, priv_tx);
@@ -500,7 +502,7 @@ int mlx5e_ktls_add_tx(struct net_device *netdev, struct sock *sk,
 
 	return 0;
 
-err_create_key:
+err_pool_push:
 	pool_push(pool, priv_tx);
 	return err;
 }

drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c

@@ -89,8 +89,8 @@ struct mlx5e_macsec_rx_sc {
 };
 
 struct mlx5e_macsec_umr {
+	u8 __aligned(64) ctx[MLX5_ST_SZ_BYTES(macsec_aso)];
 	dma_addr_t dma_addr;
-	u8 ctx[MLX5_ST_SZ_BYTES(macsec_aso)];
 	u32 mkey;
 };
 

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

@@ -4169,13 +4169,17 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev,
 		struct xsk_buff_pool *xsk_pool =
 			mlx5e_xsk_get_pool(&chs->params, chs->params.xsk, ix);
 		struct mlx5e_xsk_param xsk;
+		int max_xdp_mtu;
 
 		if (!xsk_pool)
 			continue;
 
 		mlx5e_build_xsk_param(xsk_pool, &xsk);
+		max_xdp_mtu = mlx5e_xdp_max_mtu(new_params, &xsk);
 
-		if (!mlx5e_validate_xsk_param(new_params, &xsk, mdev)) {
+		/* Validate XSK params and XDP MTU in advance */
+		if (!mlx5e_validate_xsk_param(new_params, &xsk, mdev) ||
+		    new_params->sw_mtu > max_xdp_mtu) {
 			u32 hr = mlx5e_get_linear_rq_headroom(new_params, &xsk);
 			int max_mtu_frame, max_mtu_page, max_mtu;
 
@@ -4185,9 +4189,9 @@ static bool mlx5e_xsk_validate_mtu(struct net_device *netdev,
 			 */
 			max_mtu_frame = MLX5E_HW2SW_MTU(new_params, xsk.chunk_size - hr);
 			max_mtu_page = MLX5E_HW2SW_MTU(new_params, SKB_MAX_HEAD(0));
-			max_mtu = min(max_mtu_frame, max_mtu_page);
+			max_mtu = min3(max_mtu_frame, max_mtu_page, max_xdp_mtu);
 
-			netdev_err(netdev, "MTU %d is too big for an XSK running on channel %u. Try MTU <= %d\n",
+			netdev_err(netdev, "MTU %d is too big for an XSK running on channel %u or its redirection XDP program. Try MTU <= %d\n",
 				   new_params->sw_mtu, ix, max_mtu);
 			return false;
 		}
@@ -4979,8 +4983,6 @@ void mlx5e_build_nic_params(struct mlx5e_priv *priv, struct mlx5e_xsk *xsk, u16
 	/* TX inline */
 	mlx5_query_min_inline(mdev, &params->tx_min_inline_mode);
 
-	params->tunneled_offload_en = mlx5_tunnel_inner_ft_supported(mdev);
-
 	/* AF_XDP */
 	params->xsk = xsk;
 
@@ -5285,7 +5287,7 @@ static int mlx5e_init_nic_rx(struct mlx5e_priv *priv)
 	}
 
 	features = MLX5E_RX_RES_FEATURE_PTP;
-	if (priv->channels.params.tunneled_offload_en)
+	if (mlx5_tunnel_inner_ft_supported(mdev))
 		features |= MLX5E_RX_RES_FEATURE_INNER_FT;
 	err = mlx5e_rx_res_init(priv->rx_res, priv->mdev, features,
 				priv->max_nch, priv->drop_rq.rqn,

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

@@ -755,7 +755,6 @@ static void mlx5e_build_rep_params(struct net_device *netdev)
 	mlx5e_set_rx_cq_mode_params(params, cq_period_mode);
 
 	params->mqprio.num_tc       = 1;
-	params->tunneled_offload_en = false;
 	if (rep->vport != MLX5_VPORT_UPLINK)
 		params->vlan_strip_disable = true;
 

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

@@ -3752,7 +3752,7 @@ mlx5e_clone_flow_attr_for_post_act(struct mlx5_flow_attr *attr,
 	parse_attr->filter_dev = attr->parse_attr->filter_dev;
 	attr2->action = 0;
 	attr2->counter = NULL;
-	attr->tc_act_cookies_count = 0;
+	attr2->tc_act_cookies_count = 0;
 	attr2->flags = 0;
 	attr2->parse_attr = parse_attr;
 	attr2->dest_chain = 0;
@@ -4304,6 +4304,7 @@ int mlx5e_set_fwd_to_int_port_actions(struct mlx5e_priv *priv,
 
 	esw_attr->dest_int_port = dest_int_port;
 	esw_attr->dests[out_index].flags |= MLX5_ESW_DEST_CHAIN_WITH_SRC_PORT_CHANGE;
+	esw_attr->split_count = out_index;
 
 	/* Forward to root fdb for matching against the new source vport */
 	attr->dest_chain = 0;
@@ -5304,8 +5305,10 @@ int mlx5e_tc_nic_init(struct mlx5e_priv *priv)
 	mlx5e_tc_debugfs_init(tc, mlx5e_fs_get_debugfs_root(priv->fs));
 
 	tc->action_stats_handle = mlx5e_tc_act_stats_create();
-	if (IS_ERR(tc->action_stats_handle))
+	if (IS_ERR(tc->action_stats_handle)) {
+		err = PTR_ERR(tc->action_stats_handle);
 		goto err_act_stats;
+	}
 
 	return 0;
 
@@ -5440,8 +5443,10 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv)
 	}
 
 	uplink_priv->action_stats_handle = mlx5e_tc_act_stats_create();
-	if (IS_ERR(uplink_priv->action_stats_handle))
+	if (IS_ERR(uplink_priv->action_stats_handle)) {
+		err = PTR_ERR(uplink_priv->action_stats_handle);
 		goto err_action_counter;
+	}
 
 	return 0;
 
@@ -5463,6 +5468,16 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv)
 
 void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv)
 {
+	struct mlx5e_rep_priv *rpriv;
+	struct mlx5_eswitch *esw;
+	struct mlx5e_priv *priv;
+
+	rpriv = container_of(uplink_priv, struct mlx5e_rep_priv, uplink_priv);
+	priv = netdev_priv(rpriv->netdev);
+	esw = priv->mdev->priv.eswitch;
+
+	mlx5e_tc_clean_fdb_peer_flows(esw);
+
 	mlx5e_tc_tun_cleanup(uplink_priv->encap);
 
 	mapping_destroy(uplink_priv->tunnel_enc_opts_mapping);

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

@@ -723,11 +723,11 @@ mlx5_eswitch_add_fwd_rule(struct mlx5_eswitch *esw,
 
 	flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST;
 	for (i = 0; i < esw_attr->split_count; i++) {
-		if (esw_is_indir_table(esw, attr))
-			err = esw_setup_indir_table(dest, &flow_act, esw, attr, false, &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);
+		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);
 

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

@@ -70,7 +70,6 @@ static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev,
 
 	params->packet_merge.type = MLX5E_PACKET_MERGE_NONE;
 	params->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN;
-	params->tunneled_offload_en = false;
 
 	/* CQE compression is not supported for IPoIB */
 	params->rx_cqe_compress_def = false;

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

@@ -1364,8 +1364,8 @@ static void mlx5_unload(struct mlx5_core_dev *dev)
 {
 	mlx5_devlink_traps_unregister(priv_to_devlink(dev));
 	mlx5_sf_dev_table_destroy(dev);
-	mlx5_sriov_detach(dev);
 	mlx5_eswitch_disable(dev->priv.eswitch);
+	mlx5_sriov_detach(dev);
 	mlx5_lag_remove_mdev(dev);
 	mlx5_ec_cleanup(dev);
 	mlx5_sf_hw_table_destroy(dev);
@@ -1789,11 +1789,11 @@ static void remove_one(struct pci_dev *pdev)
 	struct mlx5_core_dev *dev  = pci_get_drvdata(pdev);
 	struct devlink *devlink = priv_to_devlink(dev);
 
+	set_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state);
 	/* mlx5_drain_fw_reset() is using devlink APIs. Hence, we must drain
 	 * fw_reset before unregistering the devlink.
 	 */
 	mlx5_drain_fw_reset(dev);
-	set_bit(MLX5_BREAK_FW_WAIT, &dev->intf_state);
 	devlink_unregister(devlink);
 	mlx5_sriov_disable(pdev);
 	mlx5_crdump_disable(dev);

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

@@ -82,6 +82,16 @@ static u16 func_id_to_type(struct mlx5_core_dev *dev, u16 func_id, bool ec_funct
 	return func_id <= mlx5_core_max_vfs(dev) ?  MLX5_VF : MLX5_SF;
 }
 
+static u32 mlx5_get_ec_function(u32 function)
+{
+	return function >> 16;
+}
+
+static u32 mlx5_get_func_id(u32 function)
+{
+	return function & 0xffff;
+}
+
 static struct rb_root *page_root_per_function(struct mlx5_core_dev *dev, u32 function)
 {
 	struct rb_root *root;
@@ -665,20 +675,22 @@ static int optimal_reclaimed_pages(void)
 }
 
 static int mlx5_reclaim_root_pages(struct mlx5_core_dev *dev,
-				   struct rb_root *root, u16 func_id)
+				   struct rb_root *root, u32 function)
 {
 	u64 recl_pages_to_jiffies = msecs_to_jiffies(mlx5_tout_ms(dev, RECLAIM_PAGES));
 	unsigned long end = jiffies + recl_pages_to_jiffies;
 
 	while (!RB_EMPTY_ROOT(root)) {
+		u32 ec_function = mlx5_get_ec_function(function);
+		u32 function_id = mlx5_get_func_id(function);
 		int nclaimed;
 		int err;
 
-		err = reclaim_pages(dev, func_id, optimal_reclaimed_pages(),
-				    &nclaimed, false, mlx5_core_is_ecpf(dev));
+		err = reclaim_pages(dev, function_id, optimal_reclaimed_pages(),
+				    &nclaimed, false, ec_function);
 		if (err) {
-			mlx5_core_warn(dev, "failed reclaiming pages (%d) for func id 0x%x\n",
-				       err, func_id);
+			mlx5_core_warn(dev, "reclaim_pages err (%d) func_id=0x%x ec_func=0x%x\n",
+				       err, function_id, ec_function);
 			return err;
 		}